ALBERT

Description: The objectives of this program are as follows: modelling of dynamics of composite tubular space structure truss members; and utilization of ultrasonic waves as probes for material and defect characterization. This discussion is presented in viewgraph format.

Description: The objectives of this program are as follows: modelling of guided waves in fiber-reinforced plates in terms of different modes; and analysis of scattering by transverse cracks using modal representation. A hybrid numerical method combining the finite element representation of a region around the crack with the modal representation in the exterior region will be used in this program. Modes will be obtained using the through-the-thickness discretization of the displacement field.

Description: In the interest of aviation safety, NASA and the FAA are jointly conducting research to determine the applicability of airborne, coherent Doppler radar techniques to detect early microburst in wind shear conditions during aircraft takeoff and landing. Researchers have developed a computer model of the radar which predicts its response when viewing a simulated microburst against the simulated clutter background of an airport, the so-called radar microburst ground clutter model. Studies employing this model revealed that Doppler radar can accurately detect microburst ahead of the aircraft in time for pilot evasive response, but flight experiments will be required for complete performance evaluation of the system. An experimental X band radar is being developed for future flight experiments to verify the simulation modeling results. A description of the experimental radar, recording equipment, and its installation on the NASA 515 aircraft is presented. The flight experiments to be conducted are also described.

Description: A new technique is described for calibration of complex multipolarization SAR imagery. Scatterer reciprocity and lack of correlation between like- and cross-polarized radar echoes for natural targets are used to remove cross-polarized contamination in the radar data channels without the use of known ground targets. If known targets are available, all data channels can be calibrated relative to one another and absolutely as well. The method is verified with airborne SAR data.

Description: The SAR (synthetic-aperture radar) data-processing algorithm to be used for the Magellan mission is described. Radar system design, SAR data characteristics, and hardware (H/W) constraints, which are critical to the processing algorithm design, are highlighted. Data flow and the H/W architecture are given to show the real-time data processing capability. Simulation results obtained from processing the synthetic point-target echos are presented to demonstrate the performance of the processing algorithm.

Description: The analysis and low-power testing of a square multiflare horn designed for 1-megawatt CW operation is described. Design considerations for the five-port input section are discussed. The required aperture modes are determined from radiation pattern considerations, and analysis of the multiflare section is carried out using mode matching. Measurements demonstrate that a circularly symmetric beam is produced with relatively low sidelobes.

Description: The cross polarization characteristics of rectantular patch antennas are studied experimentally. Data are presented showing the dependence of the copolarization to cross-polarization ratio on the aspect ratio in both the E and H planes. Three substrate thicknesses are included and the variation with resonant frequency is examined.

Description: Synthesis of dual-shaped offset reflector antennas to control the exit aperture distribution of amplitude and phase has received considerable attention in recent years. For a given feed illumination and desired aperture field distribution, an exact formulation of the problem of simultaneously synthesizing the shapes of the sub and main reflectors was presented recently by Galindo-Israel et al. (1987) in terms of a set of nonlinear first-order differential equations. In this paper, a numerical approach to solving these equations is discussed which circumvents some of the difficulties encountered by Galindo-Israel et al., particularly for small values of theta.

Description: This paper considers complex transcendental eigenvalue problems where one is interested in pairs of eigenvalues that are restricted to take real values only. Such eigenvalue problems arise in dynamic stability analysis of nonconservative physical systems, i.e., flutter analysis of aeroelastic systems. Some available solution methods are discussed and a new method is presented. Two computational approaches are described for analytical evaluation of the sensitivities of these eigenvalues when they are dependent on other parameters. The algorithms presented are illustrated through examples.

Description: The technique described uses the radar return from natural targets and at least one trihedral corner reflector to calibrate compressed polarimetric radar data for relative amplitude, relative phase, absolute amplitude, and system crosstalk. The crosstalk method is based on the theoretical result that, for natural targets with azimuthal symmetry, the copolarized and cross-polarized components of the scattering matrix are uncorrelated. This method does not require any external calibration targets to be deployed. Since compressed data are used, it is necessary to model the transmitting and receiving systems as reciprocal. The method is used to estimate the crosstalk parameters of the NASA/JPL aircraft for different types of terrain and for two frequencies. For the C-band systems the crosstalk is less than -20 dB for all ranges in the images. The crosstalk of the L-band system is a function of range, however, and may be as poor as -10 dB in the near range, leading to a noticeable distortion of the polarization signatures.

Description: A technique for estimation of the Doppler centroid of an SAR in the presence of large uncertainty in antenna boresight pointing is described. Also investigated is the image degradation resulting from data processing that uses an ambiguous centroid. Two approaches for resolving ambiguities in Doppler centroid estimation (DCE) are presented: the range cross-correlation technique and the multiple-PRF (pulse repetition frequency) technique. Because other design factors control the PRF selection for SAR, a generalized algorithm is derived for PRFs not containing a common divisor. An example using the SIR-C parameters illustrates that this algorithm is capable of resolving the C-band DCE ambiguities for antenna pointing uncertainties of about 2-3 deg.

Description: A theoretical analysis of polarization filtering for the bistatic case is developed for optimum discrimination between two types of targets. The resulting method is half analytical and half numerical. Because it is based on the Stokes matrix representation, the targets of interest can be extended targets. The scattered field from such targets is partially polarized. This method is then applied to the monostatic case with numerical examples relying on the JPL (Jet Propulsion Laboratory) full-polarimetric L-band radar data. A matched filter to maximize the power ratio between urban and natural targets is developed. The results show that the same filter is optimal for both ocean and forest targets as natural targets.

Description: The unique characteristics of a spaceborne SAR (synthetic aperture radar) operating in a squint mode include large range walk and large variation in the Doppler centroid as a function of range. A pointing control technique to reduce the Doppler drift and a new processing algorithm to accommodate large range walk are presented. Simulations of the new algorithm for squint angles up to 20 deg and look angles up to 44 deg for the Earth Observing System (Eos) L-band SAR configuration demonstrate that it is capable of maintaining the resolution broadening within 20 percent and the ISLR within a fraction of a decibel of the theoretical value.

Description: A wavelength that lies within a spectral interval of reduced solar emission (a Fraunhofer line) can carry optical communications with reduced interference from direct or reflected background sunlight. Suitable Fraunhofer lines are located within the tuning range of good candidate lasers. The laser should be tunable dynamically to track Doppler shifts in the sunlight incident on any solar system body that may appear in the background as viewed by the receiver. A Fraunhofer filter used with a direct-detection receiver should be tuned to match the Doppler shifts of the source and background. The required tuning calculated here for various situations is also required if, instead, one uses a heterodyne receiver with limited post-detection bandwidth.

Description: Delamination growth caused by local buckling of a delaminated group of plies was investigated. Delamination growth was assumed to be governed by the strain energy release rates, G(1), G(2) and G(3). The strain energy release rates were calculated using a geometrically nonlinear three-dimensional finite element analysis. The program is described and several checks of the analysis are discussed. Based on a limited parametric study, the following conclusions were reached: (1) the problem is definitely mixed mode (in some cases G(1) is larger than G(2), for other cases the opposite is true); (2) in general, there is a large gradient in the strain energy release rates along the delamination front; (3) the locations of maximum G(1) and G(2) depend on the delamination shape and the applied strain; (4) the mode 3 component was negligible for all cases considered; and (5) the analysis predicted that parts of the delamination would overlap. The results presented did not impose contact constraints to prevent overlapping. Further work is needed to determine the effects of allowing the overlapping.

Description: The authors investigate microwave imaging of metallic objects using a diversity method and interpret and predict the reconstructed image from an approach based on analysis of the scattering mechanism and a back-projection algorithm used in image retrieval. The connection between the various scattering mechanisms and the reconstructed images is discussed, what the images represent is interpreted, and a prediction is made as to what the image will look like over given spectral and angular windows. A brief description is given of the microwave diversity imaging system and the formulation of the microwave diversity imaging based on the physical optics approximation. The scattering mechanism of a complex shaped metallic object is then briefly reviewed and an alternate approach to interpreting the reconstructed image based on the understanding of the scattering mechanism and the reconstruction algorithm is given. Several numerical and experimental examples are included to support this interpretation approach.

Description: An examination is presented of three techniques used for the efficient computation of fields diffracted by a subreflector that has been shaped by geometrical optics synthesis. It is found that these techniques, which are based on the geometrical theory of diffraction (GTD), produce errors in the computed fields that are specific to shaped reflectors. These errors are examined for a reflector system shaped to produce maximum gain from a tapered feed illumination. The discrepancies are directly related to the caustic being located near an observation point of the GTD calculations. The errors found are localized, and they increase in magnitude as the caustic approaches the main reflector. In a general offset geometry, the location of the caustic may be located arbitrarily close to the main reflector given a prescribed output aperture distribution. For the specific case considered here-the common situation of shaping to produce maximum gain-the caustic is located near the edge of the main reflector and on the reflection shadow boundary. A local correction is derived which creates a uniform solution through the caustic and across the reflection shadow boundary. Away from this point the calculation receeds to the standard GTD solution.

Description: The effect of filtering on the signal-to-noise ratio (SNR) of a coherently demodulated band-limited signal is determined in the presence of worst-case amplitude ripple. The problem is formulated as an optimizaton in the Hilbert space L2. The form of the worst-case amplitude ripple is specified, and the degradation in the SNR is derived in closed form. It is shown that, when the maximum passband amplitude ripple is 2Delta (peak-to-peak), the SNR is degraded by at most (1-Delta-squared), even when the ripple is unknown or uncompensated. For example, an SNR loss of less than 0.01 dB due to amplitude ripple can be assured by keeping the amplitude ripple under 0.42 dB.

Description: The propagation of elastic waves of the type induced by impact are studied for a special class of cylinders. In this class are thin finite cylinders, for which Mindlin's equations are appropriate, on which point masses are attached. An approximation is derived valid for frequencies below cutoff of the lowest shear mode. An eigenfunction expansion is used to compute the transient response of the coupled system.

Description: The method of boundary elements is adapted to the dynamics of elastic axisymmetric solids. Three scalar potentials are used, each satisfying the Helmholtz equation and each represented by its own source distribution. The sources are distributed over a surface enclosing the solid boundary. Stress or displacement conditions on the boundary are realized by third-order derivatives of the potentials. Discrete ring elements on the source enclosure combine their influences at discrete ring elements on the solid boundary.

Description: While the standard mode acceleration formulation in structural dynamics has often been interpreted to suggest that the reason for improved convergence obtainable is that the dynamic correction factor is divided by the modal frequencies-squared, an alternative formulation is presented which clearly indicates that the only difference between mode acceleration and mode displacement data recovery is the addition of a static correction term. Attention is given to the advantages in numerical implementation associated with this alternative, as well as to an illustrative example.

Description: Radar cross section (RCS) reduction by absorber covering is experimentally studied by employing microwave diversity imaging. Experimental results show that broadband absorber covering is not effective at reducing the co-polarized (the transmitting and receiving antennas have opposite sense of circular polarization) RCS of a plate when the incident wave approximates the edge-on direction but is effective at reducing the cross-polarized (both the transmitting and receiving antennas have the same sense of circular polarization) RCS for all incident directions. The surface current absorber covering is effective at reducing the nonspecular energy and multiple bounces regardless of the polarization status of the measurement.

Description: A key step in the application of formal automated design techniques to structures under transient loading is the calculation of sensitivities of response quantities to the design parameters. This paper considers response quantities to the design parameters. This paper considers structures with general forms of damping acted on by general transient loading and addresses issues of computational errors and computational efficiency. The equations of motion are reduced using the traditional basis of vibration modes and then integrated using a highly accurate, explicit integration technique. A critical point constraint formulation is used to place constraints on the magnitude of each response quantity as a function of time. Three different techniques for calculating sensitivities of the critical point constraints are presented. The first two are based on the straightforward application of the forward and central difference operators, respectively. The third is based on explicit differentiation of the equations of motion. Condition errors, finite difference truncation errors, and modal convergence errors for the three techniques are compared by applying them to a simple five-span-beam problem. Sensitivity results are presented for two different transient loading conditions and for both damped and undamped cases.

Description: A new and simple method of finite-element grid improvement is presented. The objective is to improve the accuracy of the analysis. The procedure is based on a minimization of the trace of the stiffness matrix. For a broad class of problems this minimization is seen to be equivalent to minimizing the potential energy. The method is illustrated with the classical tapered bar problem examined earlier by Prager and Masur. Identical results are obtained.

Description: Assuming a single-mode transverse displacement, a simple formula is derived for the transverse load-displacement relationship of a plate under in-plane compression. The formula is used to derive a simple analytical expression for the nonlinear dynamic response of postbuckled plates under sinusoidal or random excitation. The highly nonlinear motion of snap-through can be easily interpreted using the single-mode formula. Experimental results are obtained with buckled and cylindrical aluminum panels using discrete frequency and broadband excitation of mechanical and acoustic forces. Some important effects of the snap-through motion on the dynamic response of the postbuckled plates are described. Static tests were used to identify the deformation shape during snap-through.

Description: An experimental investigation of composite to composite bonded joints was undertaken to study the effect of bond thickness on debond growth rate under cyclic loading and critical strain energy release rate under static loading. Double cantilever beam specimens of graphite/epoxy adherends bonded with EC 3445 were tested under mode I loading. A different behavior of fracture and fatigue strength was observed with variation of bondline thickness.

Description: The gain of circularly polarized (CP) array antennas realized by proper phasing of sequentially rotated linearly polarized (LP) elements is compared to that of arrays using CP elements and demonstrated by calculations for microstrip patch elements. When element spacing is large and array size is small, the advantages of LP elements are offset by the significant reduction in gain due to high cross polarized lobes in the diagonal planes. For large arrays of closely spaced elements, this gain loss reduces to a negligible amount. However, for spacings above a critical value of about 0.7 wavelengths, unacceptably high gain losses will be incurred.

Description: The use of an imaging radar polarimeter data for unsupervised classification of scattering behavior is described by comparing the polarization properties of each pixel in a image to that of simple classes of scattering such as even number of reflections, odd number of reflections, and diffuse scattering. For example, when this algorithm is applied to data acquired over the San Francisco Bay area in California, it classifies scattering by the ocean as being similar to that predicted by the class of odd number of reflections, scattering by the urban area as being similar to that predicted by the class of even number of reflections, and scattering by the Golden Gate Park as being similar to that predicted by the diffuse scattering class. It also classifies the scattering by a lighthouse in the ocean and boats on the ocean surface as being similar to that predicted by the even number of reflections class, making it easy to identify these objects against the background of the surrounding ocean. The algorithm is also applied to forested areas and shows that scattering from clear-cut areas and agricultural fields is mostly similar to that predicted by the odd number of reflections class, while the scattering from tree-covered areas generally is classified as being a mixture of pixels exhibiting the characteristics of all three classes, although each pixel is identified with only a single class.

Description: The electromagnetic properties of a structure that is both chiral and periodic are investigated using coupled-mode equations. The periodicity is described by a sinusoidal perturbation of the permittivity, permeability, and chiral admittance. The coupled-mode equations are derived from physical considerations and used to examine bandgap structure and reflected and transmitted fields. Chirality is observed predominantly in transmission, whereas periodicity is present in both reflection and transmission.

Description: A method is presented for analyzing a finite planar array of circular microstrip patches fed by coaxial probes. The self- and mutual impedances between array elements are calculated using the method of moments with the dyadic Green's function for a dielectric layer on a ground plane. The patch circuits are determined by using the reaction integral equation. The active input impedance as well as the active element pattern of the array are computed from a knowledge of the resultant patch currents. The calculated results for two-element and eight-element linear arrays are in good agreement with experimental data. The active reflection coefficient and element pattern for the center and edge elements of a two-dimensional array as a function of scan angle are also presented.

Description: The issue of reducing the cost of phased array vehicle antennas through the use of a lens feeding arrangement instead of phase shifters at each element is addressed. In particular, the economic viability of a mobile satellite system (MSAT) is largely dependent on the efficient use of the allocated scarce spectrum and orbit as well as the satellite power. the type of vehicle antenna used will play a critical role in achieving this efficiency. A standard design approach for an electronically steered array uses phase shifters at each element to provide beam steering. A method for reducing the required number of phase shifters by using an R-KR lens feed network is outlined. The authors briefly discuss the phase shifter approach to beam steering, examine various lens feed techniques, and describe the R-KR lens approach. The lens feed network architecture is examined, a computer model for simulation of the array is presented, and the results of analysis of a suggested design for the MSAT application are given. In addition, satellite acquisition and tracking considerations are investigated.

Description: Slot timing recovery in a direct-detection optical PPM communication system can be achieved by processing the photodetector output waveform with a nonlinear device whose output forms the input to a phase-locked loop. The choice of a simple transition detector as the nonlinearity is shown to give satisfactory synchronization performance. The rms phase error of the recovered slot clock and the effect of slot timing jitter on the bit error probability were directly measured. The experimental system consisted of an AlGaAs laser diode (wavelength = 834 nm) and a silicon avalanche photodiode photodetector. The system used Q = 4 PPM signaling and operated at a source data rate of 25 Mbits/s. The mathematical model developed to compute the rms phase error of the recovered clock is shown to be in good agreement with results of actual measurements of phase errors. The use of the recovered slot clock in the receiver resulted in no significant degradation in receiver sensitivity compared to a system with perfect slot timing. The system achieved a bit error probability of 10 to the -6th at a received optical signal energy of 55 detected photons per information bit.

Description: The feasibility of using a viscoplastic model developed by Robinson to perform a nonlinear structural analysis was investigated. The paper presents analytical solutions for three classical problems: (1) a pressurized thick-walled cylinder, (2) a thin rotating disk, and (3) a pressurized thick-walled sphere. The analytical expressions derived for the stress and the strain rates for these components are general in nature as they consider both the mechanical and thermal loadings to be time-dependent. A computer program VISTAN (VIscoplastic STructural ANalyzer) was developed to obtain the stress and strain distributions. The finite element solutions for these problems are also presented. The numerical results pertaining to isothermal loading conditions are provided. The results obtained demonstrate the feasibility of using the viscoplastic model developed by Robinson to perform nonlinear structural analyses.

Description: A combined experimental and analytical study of bearing and bypass loading on single-fastener specimens of a 16-ply, quasi-isotropic T300/5208 graphite/epoxy laminate with a centrally located hole is reported. The specimens were loaded in either tension or compression, and onset damage, ultimate strengths, and corresponding failure modes were determined. The tension data showed the expected linear interaction for combined bearing-bypass loading with damage developing in the net-section tension mode. However, the bearing-onset strengths showed an unexpected interaction of the bearing and compressive bypass loads in which the latter reduced the bearing-onset strength. A linear finite element analysis showed that bearing-bypass loading had a marked influence on the bolt-hole contact which in turn had a significant effect on local stresses.

Description: A finite element formulation is presented for determining the large-amplitude free and steady-state forced vibration response of arbitrarily laminated anisotropic composite rectangular thin plates. The nonlinear stiffness and harmonic force matrices of an arbitrarily laminated composite rectangular plate element are developed for nonlinear free and forced vibration analyses. The linearized updated-mode method with nonlinear time function approximation is employed for the solution of the system nonlinear eigenvalue equations. The amplitude-frequency relations for convergence with gridwork refinement, different boundary conditions, aspect ratios, lamination angles and number of plies are presented. The finite element results are compared with available approximate continuum solutions.

Description: The objectives are to present the derivation of the new virtual crack closure technique, evaluate the accuracy of the technique, and finally to present the results of a limited parametric study of laminates with a postbuckled delamination. Although the new virtual crack closure technique is general, only homogeneous, isotropic laminates were analyzed. This was to eliminate the variation of flexural stiffness with orientation, which occurs even for quasi-isotropic laminates. This made it easier to identify the effect of geometrical parameters on G. The new virtual crack closure technique is derived. Then the specimen configurations are described. Next, the stress analyses is discussed. Finally, the virtual crack closure technique is evaluated and then used to calculate the distribution of G along the delamination front of several laminates with a postbuckled delamination.

Description: The moment method solution to the problem of a reactively loaded circular patch is presented. Using the reaction integral equation in conjuction with the method of moments, parameters of the Thevenin's equivalent network for the loaded patch are obtained. From the equivalent network parameters an expression for the imput impedance of the loaded patch is derived. A design procedure for a circularly polarized disk antenna is presented. Computed results are compared with the experimental data.

Description: A description of the finite element implementation of Robinson's unified viscoplastic model into the General Purpose Finite Element Program (MARC) is presented. To demonstrate its application, the implementation is applied to some uniaxial and multiaxial problems. A comparison of the results for the multiaxial problem of a thick internally pressurized cylinder, obtained using the finite element implementation and an analytical solution, is also presented. The excellent agreement obtained confirms the correct finite element implementation of Robinson's model.

Description: The use of commercially available 1-bit LSI correlator chips whose main function is the correlation of sequences of plus ones and minus ones (rather than zeros and ones) as building blocks in the design of multibit correlators is investigated. A radix-2 number system in which the bit values are + or - 1 (no zero) is discussed. The development of a simple conversion algorithm to find the + or - 1 representation of a number given in 2's complement is presented. Some of the issues raised by this conversion algorithm are discussed. It is shown that, if the analog signal is available, there is a simple step to take before the analog-to-digital conversion in order to improve the overall precision. The construction of multibit correlators based on the + or - 1 system is compared to alternative approaches.

Description: In the dynamic formulation of holonomic and nonholonomic systems based on D'Alembert-Lagrange equation, the forces of constraints are maintained in the dynamic equations by introducing auxiliary variables, called Lagrange multipliers. This approach introduces a set of generalized reaction forces associated with the system generalized coordinates. Different sets of variables can be used as generalized coordinates and accordingly, the generalized reactions associated with these generalized coordinates may not be the actual reaction forces at the joints. In rigid body dynamics, the generalized reaction forces and the actual reaction forces at the joints represent equipollent systems of forces since they produce the same total forces and moments at and about any point on the rigid body. This is not, however, the case in deformable body analyses wherein the generalized reaction forces depend on the system generalized reference and elastic coordinates. In this paper, a method for determining the actual reaction forces at the joints from the generalized reaction forces in deformable multibody systems is presented.

Description: The Boundary Force Method (BFM), a form of indirect boundary element method, is used to analyze composite laminates with cracks. The BFM uses the orthotropic elasticity solution for a concentrated horizontal and vertical force and a moment applied at a point in a cracked, infinite sheet as the fundamental solution. The necessary stress functions for this fundamental solution were formulated using the complex variables theory of orthotropic elasticity. The current method is an improvement over a previous method using only forces and no moment. The improved method was verified by comparing it to accepted solutions for a finite-width, center-crack specimen subjected to uniaxial tension. Four graphite/epoxy laminates were used: (0 + or - 45/90)sub s, (0), (+ or - 45)sub s, and (+ or - 30)sub s. The BFM results agreed well with accepted solutions. Convergence studies showed that with the addition of the moment in the fundamental solution, the number of boundary elements required for a converged solution was significantly reduced. Parametric studies were done for two configurations for which no orthotropic solutions are currently available; a single edge crack and an inclined single edge crack.

Description: The fundamental relationship between the morphology of a composite laminate and the resulting free edge effects is explored and related to delamination failures. Cross-ply, angle-ply, and quasi-isotropic laminates are discussed in detail. It is shown that the local mismatch in elastic properties of adjacent layers and the global stacking sequence of a laminate both have a significant influence on the interlaminar stresses and delamination failures.

Description: A method is developed for the optimal design of composite links based on dynamic performance criteria directly related to structural modal damping and dynamic stiffness. An integrated mechanics theory correlates structural composite damping to the parameters of basic composite material systems, laminate parameters, link shape, and modal deformations. The inclusion of modal properties allows the selective minimization of vibrations associated with specific modes. Ply angles and fiber volumes are tailored to obtain optimal combinations of damping and stiffness. Applications to simple composite links indicate wide margins for trade-offs and illustrate the importance of various design variables to the optimal design.

Description: Local buckling can cause large interlaminar stresses along the delamination front, which can lead to delamination growth. This paper examines several methods of calculating strain-energy release rates, which are often used to predict delamination growth. The thin-film plate analysis, which was least expensive, calculated the total strain-energy release rate G(T) quite accurately. However, the stress field along the delamination front is highly mixed-mode and has no fixed ratio of G(I) to G(II). Since plate analysis can only calculate G(T), it would not be useful for accurate predictions of delamination growth if mode mix is important.

Description: Over the years, NASA has been conducting the Advanced Earth-to-Orbit (AETO) Propulsion Technology Program to provide the knowledge, understanding, and design methodology that will allow the development of advanced Earth-to-orbit propulsion systems with high performance, extended service life, automated operations, and diagnostics for in-flight health monitoring. The objective of the Aerothermodynamic Loads Definition Study is to develop methods to more accurately predict the operating environment in AETO propulsion systems, such as the Space Shuttle Main Engine (SSME) powerhead. The approach taken consists of 2 parts: to modify, apply, and disseminate existing computational fluid dynamics tools in response to current needs and to develop new technology that will enable more accurate computation of the time averaged and unsteady aerothermodynamic loads in the SSME powerhead. The software tools are detailed. Significant progress was made in the area of turbomachinery, where there is an overlap between the AETO efforts and research in the aeronautical gas turbine field.

Description: The purpose of the Probabilistic Structural Analysis Method (PSAM) project is to develop structural analysis capabilities for the design analysis of advanced space propulsion system hardware. The boundary element method (BEM) is used as the basis of the Probabilistic Advanced Analysis Methods (PADAM) which is discussed. The probabilistic BEM code (PBEM) is used to obtain the structural response and sensitivity results to a set of random variables. As such, PBEM performs analogous to other structural analysis codes such as finite elements in the PSAM system. For linear problems, unlike the finite element method (FEM), the BEM governing equations are written at the boundary of the body only, thus, the method eliminates the need to model the volume of the body. However, for general body force problems, a direct condensation of the governing equations to the boundary of the body is not possible and therefore volume modeling is generally required.

Description: A major research and technology program in Probabilistic Structural Analysis Methods (PSAM) is currently being sponsored by the NASA Lewis Research Center with Southwest Research Institute as the prime contractor. This program is motivated by the need to accurately predict structural response in an environment where the loadings, the material properties, and even the structure may be considered random. The heart of PSAM is a software package which combines advanced structural analysis codes with a fast probability integration (FPI) algorithm for the efficient calculation of stochastic structural response. The basic idea of PAAM is simple: make an approximate calculation of system response, including calculation of the associated probabilities, with minimal computation time and cost, based on a simplified representation of the geometry, loads, and material. The deterministic solution resulting should give a reasonable and realistic description of performance-limiting system responses, although some error will be inevitable. If the simple model has correctly captured the basic mechanics of the system, however, including the proper functional dependence of stress, frequency, etc. on design parameters, then the response sensitivities calculated may be of significantly higher accuracy.

Description: Two-dimensional optical strain measurements on high temperature test specimens are presented. This two-dimensional capability is implemented through a rotatable sensitive strain axis. Three components of surface strain can be measured automatically, from which the first and second principal strains are calculated. One- and two-dimensional strain measurements at temperatures beyond 750 C with a resolution of 15 microstrain are demonstrated. The system is based on a one-dimensional speckle shift technique. The speckle shift technique makes use of the linear relationship between surface strain and the differential shift of laser speckle patterns in the diffraction plane. Laser speckle is a phase effect that occurs when spatially coherent light interacts with an optically rough surface. Since speckle is generated by any diffusely reflecting surface, no specimen preparation is needed to obtain a good signal. Testing was done at room temperature on a flat specimen of Inconel 600 mounted in a fatigue testing machine. A load cell measured the stress on the specimen before and after acquiring the speckle data. Strain components were measured at 0 C (parallel to the load axis) and at plus or minus 45 C, and plots indicate the calculated values of the first and second principal strains. The measured values of Young's modulus and Poisson's ratio are in good agreement with handbook values. Good linearity of the principal strain moduli at high temperatures indicate precision and stability of the system. However, a systematic error in the high-temperature test setup introduced a scale factor in the slopes of the two-dimensional stress-strain curves. No high temperature effects, however, have been observed to degrade speckle correlation.

Description: The fusion of the probabilistic finite element method (PFEM) and reliability analysis for probabilistic fracture mechanics (PFM) is presented. A comprehensive method for determining the probability of fatigue failure for curved crack growth was developed. The criterion for failure or performance function is stated as: the fatigue life of a component must exceed the service life of the component; otherwise failure will occur. An enriched element that has the near-crack-tip singular strain field embedded in the element is used to formulate the equilibrium equation and solve for the stress intensity factors at the crack-tip. Performance and accuracy of the method is demonstrated on a classical mode 1 fatigue problem.

Description: The probabilistic structural analysis method (PSAM) was developed to analyze the effects of fluctuating loads, variable material properties, and uncertain analytical models especially for high performance structures such as the Space Shuttle Main Engine turbopump blades. Risk is calculated after expensive service experience. However, probabilistic structural analysis provides a rational alternative method to quantify uncertainties in the structural performance and durability. NESSUS (Numerical Evaluation of Stochastic Structures Under Stress) was developed as a probabilistic structural analysis computer code which integrates finite element methods and reliability algorithms, capable to predicting the probability distributions of structural response variables such as stress, displacement, natural frequencies, and buckling loads. This computer code is detailed.

Description: The Probabilistic Structural Analysis Methods (PSAM) project developed at the Southwest Research Institute integrates state-of-the-art structural analysis techniques with probability theory for the design and analysis of complex large-scale engineering structures. An advanced efficient software system (NESSUS) capable of performing complex probabilistic analysis has been developed. NESSUS contains a number of software components to perform probabilistic analysis of structures. These components include: an expert system, a probabilistic finite element code, a probabilistic boundary element code and a fast probability integrator. The NESSUS software system is shown. An expert system is included to capture and utilize PSAM knowledge and experience. NESSUS/EXPERT is an interactive menu-driven expert system that provides information to assist in the use of the probabilistic finite element code NESSUS/FEM and the fast probability integrator (FPI). The expert system menu structure is summarized. The NESSUS system contains a state-of-the-art nonlinear probabilistic finite element code, NESSUS/FEM, to determine the structural response and sensitivities. A broad range of analysis capabilities and an extensive element library is present.

Description: Constitutive and life prediction models are developed and verified for materials typically used in hot gas path components of reusable space propulsion systems over the range of relevant operating environments. The efforts were centered on the development of crack initiation life prediction methods, while the efforts of a counterpart group were centered on the development of cyclic crack propagation life prediction methods. The complexion of the active tasks are presented. A significant new task started this year will incorporate the various material constitutive and life prediction models developed in this program into a comprehensive creep-fatigue damage analysis and life assessment computer code. The program will function as a postprocessor to general structural analysis programs (such as finite element or boundary element codes) using the output of such analyses (stress, strain, and temperature fields as functions of time) as the input to the damage analysis and life assessment code. The code will be designed to execute on engineering/scientific workstations and will feature a windowing, mouse-driven user interface. Current plans call for the code to be finished and made available for use in mid 1991.

Description: It is appropriate to note that 1988 marks the 100th anniversary of the birth of the reflector antenna. It was in 1888 that Heinrich Hertz constructed the first one, a parabolic cylinder made of sheet zinc bent to shape and supported by a wooden frame. Hertz demonstrated the existence of the electromagnetic waves that had been predicted theoretically by James Clerk Maxwell some 22 years earlier. In the 100 years since Hertz's pioneering work the field of electromagnetics has grown explosively: one of the technologies is that of remote sensing of planet Earth by means of electromagnetic waves, using both passive and active sensors located on an Earth Science Geostationary Platform (ESEP). For these purposes some exquisitely sensitive instruments were developed, capable of reaching to the fringes of the known universe, and relying on large reflector antennas to collect the minute signals and direct them to appropriate receiving devices. These antennas are electrically large, with diameters of 3000 to 10,000 wavelengths and with gains approaching 80 to 90 dB. Some of the reflector antennas proposed for ESGP are also electrically large. For example, at 220 GHz a 4-meter reflector is nearly 3000 wavelengths in diameter, and is electrically quite comparable with a number of the millimeter wave radiotelescopes that are being built around the world. Its surface must meet stringent requirements on rms smoothness, and ability to resist deformation. Here, however, the environmental forces at work are different. There are no varying forces due to wind and gravity, but inertial forces due to mechanical scanning must be reckoned with. With this form of beam scanning, minimizing momentum transfer to the space platform is a problem that demands an answer. Finally, reflector surface distortion due to thermal gradients caused by the solar flux probably represents the most challenging problem to be solved if these Large Space Antennas are to achieve the gain and resolution required of them.

Description: During a period spanning more than 5 years, low elevation radar measurements of rain were systematically obtained in the mid-Atlantic coast of the U.S. Drop size distribution measurements with a disdrometer were also acquired on the same rain days. The drop size data were utilized to convert the radar reflectivity factors to estimated rain rates for the respective rain days of operation. Applying high level algorithms to the rain data, core values of rain intensities were identified (peak rain rates), and families of rain rate isopleths analyzed. In particular, equicircle diameters of the family of isopleths enveloping peak rain intensities were statistically characterized. The presented results represents the analysis of two rain days, 12 radar scans, corresponding to 430 culled rain rate isopleths from an available data base of 22,000 contours, approximately 100 scans encompassing 17 rain days. The results presented show trends of the average rain rate vs. contour scale dimensions, and cumulative distributions of rain cell dimensions which belong to core families of precipitation.

Description: Propagation effects in the aeronautical mobile-satellite service differ from those in the fixed-satellite service and other mobile-satellite services because: small antennas are used on aircraft, and the aircraft body may affect the performance of the antenna; high aircraft speeds cause large Doppler spreads; aircraft terminals must accommodate a large dynamic range in transmission and reception; and due to their high speeds, banking maneuvers, and three-dimensional operation, aircraft routinely require exceptionally high integrity of communications, making even short-term propagation effects very important. Data and models specifically required to characterize the path impairments are discussed, which include: tropospheric effects, including gaseous attenuation, cloud and rain attenuation, fog attenuation, refraction and scintillation; surface reflection (multipath) effects; ionospheric effects such as scintillation; and environmental effects (aircraft motion, sea state, land surface type). Aeronautical mobile-satellite systems may operate on a worldwide basis, including propagation paths at low elevation angles. Several measurements of multipath parameters over land and sea were conducted. In some cases, laboratory simulations are used to compare measured data and verify model parameters. The received signals is considered in terms of its possible components: a direct wave subject to atmospheric effects, and a reflected wave, which generally contains mostly a diffuse component.

Description: In 1992, an Engineering Test Satellite 6 is scheduled to be launched by an H-2 rocket. The missions of ETS-6 are to establish basic technologies of inter-satellite communications using S-band, millimeter waves and optical beams and of fixed and mobile satellite communications using multibeam antenna on board the satellite. A plan of the experiments is introduced.

Description: The JPL Channel Simulator lab was modified to allow full duplex links and to allow the use of field propagation data for link fading. This capability will be used to test equipment for the joint AUSSAT/NASA mobile satellite experiment in July 1989.

Description: Propagation experiments for maritime, aeronautical, and land mobile satellite communications were performed using Engineering Test Satellite-Five (ETS-5). The propagation experiments are one of major mission of Experimental Mobile Satellite System (EMSS) which is aimed for establishing basic technology for future general mobile satellite communication systems. A brief introduction is presented for the experimental results on propagation problems of ETS-5/EMSS.

Description: Land-mobile satellite propagation measurements were implemented at L Band (1.5 GHz) in South-Eastern Australia during an 11 day period in October 1988. Transmissions (CW) from both the Japanese ETS-5 and INMARSAT Pacific geostationary satellites were accessed. Previous measurements in this series were performed at both L Band (1.5 GHz) and UHF (870 MHz) in Central Maryland, North-Central Colorado, and the southern United States. The objectives of the Australian campaign were to expand the data base acquired in the U.S. to another continent, to validate a U.S. derived empirical model for estimating the fade distribution, to establish the effects of directive antennas, to assess the isolation between co- and cross-polarized transmissions, to derive estimates of fade as well as non-fade durations, and to evaluate diversity reception. All these objectives were met.

Description: Results from the most recent MSAT-X field experiments, the Tower-3 Experiment and the JPL/FAA/INMARSAT MARECS-B2 Satellite Experiment, are presented. Results that distinguish the propagation environment of the tower set-up are given and explained. The configuration and flight variables of the aeronautical experiment which used an FAA aircraft and an INMARSAT satellite are described. Results that highlight the disturbances on the aeronautical satellite channel are presented. The roles of satellite-induced signal variations and of multipatch are identified and their impact on the link is discussed.

Description: Future communication systems will move toward the 20 to 30 GHz frequency range for wider bandwidth and reduced interference, and will use small earth terminals. The ESA satellite OLYMPUS is scheduled for launch this June and will be ready for use in October. OLYMPUS has 12, 20, and 30 GHz beacons. Virginia Tech and Michigan Tech are working with NASA/JPL on an OLYMPUS experiment and hardware development program. OLYMPUS beacons provide coverage of the east coast of the US sufficient for attenuation measurements. The planned hardware for the OLYMPUS experiments is illustrated. The components of the experiment are summarized. A block diagram level overview of a typical channel is given and one of the four RF front ends is shown.

Description: ACTS counters fading by resource sharing between the users. It provides a large margin only for those terminals which are at risk by unfavorable atmospheric conditions. ACTS, as an experimental satellite, provides a 5 dB clear weather margin and 10 dB additional margin via rate reduction and encoding. For the uplink, this margin may be increased by exercising uplink power control. Some of the challenges faced by the radiowave propagation community are listed. The issue of needs for the satellite are listed, both general and specific.

Description: Some information is given on the ITALSAT millimetric waves propagation experiment, which is to be conducted with the ITALSAT satellite, whose launch is foreseen for the middle of 1990. The purpose of the experiment is one of experimenting with advanced technologies and techniques employing the 20/30 GHz bands in wideband telecommunications. Among the most qualified features of this system are the multispot antenna and the exchange function performed directly onboard. Details of the experiment are given.

Description: Ground based radiometer measurements at 20.6, 31.65, and 90.0 GHz were analyzed to provide attenuation statistics, thus extending the data base of previous NAPEX studies. Using data from colocated radiosondes, comparisons of measurements and calculations of brightness temperatures are presented. The oxygen absorption model of Rosenkranz and the water vapor absorption models of Liebe and of Waters are used. Data from July 1987 at San Nicolas Island, California and from December 1987, August and November 1988 at Denver, Colorado, are included. Joint attenuation statistics at 20.6 and 31.65 GHz are presented for two locations of the Colorado Research Network for December 1987 and August 1988.

Description: The need to develop accurate models for secondary statistics of fading land mobile satellite signals has motivated a study of fading signal autocorrelations and multipath spectrum. Results of autocorrelations and power spectral densities from measured data are presented and comparisons to multipath spectrum models are made.

Description: In structural analysis the amount of computational time necessary for a solution is proportional to the number of degrees of freedom times the bandwidth squared. In implicit time analysis, this must be done at each discrete point in time. If, in addition, the problem is nonlinear, then this solution must be iterated at each point in time. If the bandwidth is large, the size of the problem that can be analyzed is severely limited. The multi-grid method is a possible algorithm which can make this solution much more computationally efficient. This method has been used for years in computational fluid mechanics. It works on the fact that relaxation is very efficient on the high frequency components of the solution (nearest neighbor interactions) and not very good on low frequency components of the solution (far interactions). The multi-grid method is then to relax the solution on a particular model until the residual stops changing. This indicates that the solution contains the higher frequency components. A coarse model is then generated for the lower frequency components to the solution. The model is then relaxed for the lower frequency components of the solution. These lower frequency components are then interpolated to the fine model. In computational fluid mechanics the equations are usually expressed as finite differences.

Description: The deployment, assembly and mission-oriented maneuvering of space structures in orbit will trigger large motions of flexible, truss-type structures. In addition, the presence of on-board controls both for attitude stabilization and specified vibration tolerance requirements may further complicate the dynamic behavior of the orbiting structures. Because of safety and cost considerations, the dynamic response of the combined structural and control systems must be predicted reliably. This need can only be met through the development of reliable and efficient simulation capabilities, since there is general agreement that on-orbit experiments should be limited because of cost, time and facility constraints. The long-term objective of this research effort is to develop a next-generation computer simulator for the dynamics and control of large space structures. The simulator will be based on integrating four research thrusts: a new multibody dynamics formulation methodology, modeling capabilities in long/slender truss-beam components with realistic joints, efficient computational procedures that can be implemented either in sequential or concurrent computers, and prototype simulation modules that can be easily processed into a modern large-scale engineering software system such as the NASA/Computational Structural Mechanics (CSM) testbed.

Description: The 3-D inelastic analysis method is a focused program with the objective to develop computationally effective analysis methods and attendant computer codes for three-dimensional, nonlinear time and temperature dependent problems present in the hot section of turbojet engine structures. Development of these methods was a major part of the Hot Section Technology (HOST) program over the past five years at Lewis Research Center.

Description: A summary of the status of this five-year project which is now in its third year of research and development is presented. The goal of the project is the development of several methodologies for probabilistic structural modeling. Probabilistic structural modeling consists of stochastic models of material properties, part geometries, boundary conditions, as well as loading conditions. The current presentation focuses on one methodology - coupling of an advanced finite element structural analysis code with probabilistic modeling strategies. The essential algorithm developments for combining the finite element and probabilistic analysis methods are reported. The validity of the resulting probabilistic structural analysis method is confirmed through a series of test problems with exact results based on Monte Carlo simulations. Additionally, the applicability of the method to a Space Propulsion System (a turbine blade) is demonstrated for static stresses.

Description: This project supports the conversion of codes in Computational Structural Mechanics (CSM) to a parallel form which will efficiently exploit the computational power available from multiprocessors. The work is a part of a comprehensive, FORTRAN-based system to form a basis for a parallel version of the NICE/SPAR combination which will form the CSM Testbed. The software is macro-based and rests on the force methodology developed by the principal investigator in connection with an early scientific multiprocessor. Machine independence is an important characteristic of the system so that retargeting it to the Flex/32, or any other multiprocessor on which NICE/SPAR might be imnplemented, is well supported. The principal investigator has experience in producing parallel software for both full and sparse systems of linear equations using the force macros. Other researchers have used the Force in finite element programs. It has been possible to rapidly develop software which performs at maximum efficiency on a multiprocessor. The inherent machine independence of the system also means that the parallelization will not be limited to a specific multiprocessor.

Description: A new numerical algorithm for the solution of large-order eigenproblems typically encountered in linear elastic finite element systems is presented. The architecture of parallel processing is utilized in the algorithm to achieve increased speed and efficiency of calculations. The algorithm is based on the frontal technique for the solution of linear simultaneous equations and the modified subspace eigenanalysis method for the solution of the eigenproblem. Assembly, elimination and back-substitution of degrees of freedom are performed concurrently, using a number of fronts. All fronts converge to and diverge from a predefined global front during elimination and back-substitution, respectively. In the meantime, reduction of the stiffness and mass matrices required by the modified subspace method can be completed during the convergence/divergence cycle and an estimate of the required eigenpairs obtained. Successive cycles of convergence and divergence are repeated until the desired accuracy of calculations is achieved. The advantages of this new algorithm in parallel computer architecture are discussed.

Description: The overall objective of this research is to develop efficient methods for the solution of linear and nonlinear systems of equations on parallel and supercomputers, and to apply these methods to the solution of problems in structural analysis. Attention has been given so far only to linear equations. The methods considered for the solution of the stiffness equation Kx=f have been Choleski factorization and the conjugate gradient iteration with SSOR and Incomplete Choleski preconditioning. More detail on these methods will be given on subsequent slides. These methods have been used to solve for the static displacements for the mast and panel focus problems in conjunction with the CSM testbed system based on NICE/SPAR.

Description: The background for the CSM Testbed Architecture and a description of the three architecture related tasks under the NASA/Langley CSM contract (NAS1-18444) are given. For each task the objectives, subtasks chosen to achieve the objectives, and the accomplishments are presented. For Task 2, Near-Term Enhancements to the CSM Testbed Architecture, the primary objectives are to quickly correct: (1) inefficiency in the GAL data manager, and (2) deficiency in the CLIP-interpreted command language, CLAMP. The corresponding modifications should preserve upward-compatibility to a reasonable extent, while at the same time increasing the flexibility and extensibility of the CSM Testbed. Researchers have increased the efficiency of GAL by a factor of 2+ and have made several improvements in CLIP. For Task 5, Matrix Algebra Methods and Utilities, the goal is to investigate the current capabilities of the CSM Testbed and the required improvements to the CSM Testbed for performing the matrix algebraic operations required for the analysis of present and future CSM focal problems using advanced methods. Researchers have made an extensive study of the matrix algebra functions in SPAR that includes documenting these routines and supplying inline comments.

Description: The PISCES project started in 1984 under the sponsorship of the NASA Computational Structural Mechanics (CSM) program. A PISCES 1 programming environment and parallel FORTRAN were implemented in 1984 for the DEC VAX (using UNIX processes to simulate parallel processes). This system was used for experimentation with parallel programs for scientific applications and AI (dynamic scene analysis) applications. PISCES 1 was ported to a network of Apollo workstations by N. Fitzgerald.

Description: In January 1987, SPARTA received a Phase I SBIR award from the NASA Lewis Research Center to investigate the feasibility of a finite element solver implemented on multiple VLSI processors. The transputer was chosen as the processor for the feasibility study since it combined low cost with high performance and was specifically designed to directly link with other transputers to form networks of multiple processors. A brief description of transputers, a summary of the SBIR feasibility study, and a discussion of issues concerning a large scale transputer based finite element solver (TBFES) and the performance levels which can be expected are discussed.

Description: Beam-waveguide (BWG) antennas provide multiple frequency band operations and other operational benefits for large ground-station antennas. Present design practices use diffraction analyses that ignore the presence of the BWG enclosure and may be inaccurate at lower frequency bands for ground-station antennas operating over multiple frequency bands. Introduced here is a new analysis approach that considers the presence of the BWG enclosure. Results based on the new analysis have revealed new understandings of the performance degradation mechanisms in a BWG antenna and have provided direction for potential design improvements.

Description: The benefits of improved radiometric tracking data have been studied for planetary approach within the inner Solar System using the Mars Rover Sample Return trajectory as a model. It was found that the benefit of improved data to approach and encounter navigation was highly dependent on the a priori uncertainties assumed for several non-estimated parameters, including those for frame-tie, Earth orientation, troposphere delay, and station locations. With these errors at their current levels, navigational performance was found to be insensitive to enhancements in data accuracy. However, when expected improvements in these errors are modeled, performance with current-accuracy data significantly improves, with substantial further improvements possible with enhancements in data accuracy.

Description: It is demonstrated how the use of interleaving/deinterleaving in trellis-coded modulation (TCM) systems can reduce the signal-to-noise ratio loss due to imperfect carrier demodulation references. Both the discrete carrier (phase-locked loop) and suppressed carrier (Costas loop) cases are considered and the differences between the two are clearly demonstrated by numerical results. These results are of great importance for future communication links to the Deep Space Network (DSN), especially from high Earth orbiters, which may be bandwidth limited.

Description: Thermal measurements of microwave transmitter feedhorn windows were performed using an imaging infrared radiometer. The measurement technique is described and results are presented for windows made of 0.001-in. Kapton (trademark of Dupont Chemical Co.) and 0.1-in. HTP-6 (Space Shuttle tile material). Measured and calculated temperatures agree well.

Description: The design and testing of the prototype horn for the proposed 1-megawatt radar are described. The unique features of this square horn include a multiflare design in which flare angle changes rather than corrugations are used to generate the required higher-order modes. A five-port combining section is used at the input. The design of this section and the multiflare section are described. Measured radiation patterns are in good agreement with theoretical patterns.

Description: Aperture efficiency measurements made during 1988 on the three 70-m stations (DSS-14, DSS-43, and DSS-63) at X-band (8420 MHz) and S-band (2295 MHz) have been analyzed and reduced to yield best estimates of antenna gain versus elevation. The analysis has been carried out by fitting the gain data to a theoretical expression based on the Ruze formula. Newly derived flux density and source-size correction factors for the natural radio calibration sources used in the measurements have been used in the reduction of the data. Peak gains measured at the three stations were 74.18 (plus or minus 0.10) dBi at X-band, and 63.34 (plus or minus 0.03) dBi at S-band, with corresponding peak aperture efficiencies of 0.687 (plus or minus 0.015) and 0.762 (plus or minus 0.006), respectively. The values quoted assume no atmosphere is present, and the estimated absolute accuracy of the gain measurements is approximately plus or minus 0.2 dB at X-band and plus or minus 0.1 dB at S-band (1-sigma values).

Description: Damage assessment of structural assemblies is treated as an identification problem. A brief review of identification methods is first presented with particular focus on the output error approach. The use of numerical optimization methods in identifying the location and extent of damage in structures is studied. The influence of damage on eigenmode shapes and static displacements is explored as a means of formulating a measure of damage in the structure. Preliminary results obtained in this study are presented and special attention is directed at the shortcomings associated with the nonlinear programming approach to solving the optimization problem.

Description: Within the past several years, the Jet Propulsion Laboratory has designed and built major ground antenna structures in Spain, Australia, and California. One of the antennas at each location is a 70 meter-diameter structure that is a retrofit of the existing 64 meter antenna. The 64 meter existing antennas were first stripped back to a 34 meter interior and then completely new construction with deeper trusses was added to extend the interior to 70 meters. The 70 meter project included the rare opportunity to collect field data to compare with predictions of the finite-element analytical models. The new quadripod design was tested for its lower mode natural frequencies and the main reflector was measured by theodolite to determine deflections of subsets of the backup-structure deformations under load. The emphasis here is to examine measurement results and possibly provide some appreciation of the relationship of predictions made from the design model to actual measurements.

Description: The application of a generalized optimality criteria to framed structures is presented. The optimality conditions, Lagrangian multipliers, resizing algorithm, and scaling procedures are all represented as a function of the objective and constraint functions along with their respective gradients. The optimization of two plane frames under multiple loading conditions subject to stress, displacement, generalized stiffness, and side constraints is presented. These results are compared to those found by optimizing the frames using a nonlinear mathematical programming technique.

Description: Recent developments in multilevel optimization are briefly reviewed. The general nature of the multilevel design task, the use of approximations to develop and solve the analysis design task, the structure of the formal multidiscipline optimization problem, a simple cantilevered beam which demonstrates the concepts of multilevel design and the basic mathematical details of the optimization task and the system level are among the topics discussed.

Description: A two-level design optimization metholology is described. A progress report of its application to Printed Wiring Board (PWB) assembly examples is given. The design of PWB assemblies is a complex task which is generally conducted as a sequential process. Individual PWBs are usually designed first, followed by the composition of the PWBs into an assembly. As a result, optimizing design considerations such as assembly reliability cannot be accomplished. This study showed that a two-level decomposition method can be employed to optimize for reliability at both the PWB- and the assembly-level in a coupled manner. The two-level decomposition method also resolved the mixed-integer nonlinear programming nature of the problem rather easily.

Description: Independent Modal Space Control (IMSC) is a technique that is often used for the control of large order structural systems. The pertinent optimization problem in the simultaneous design and control of structures is a min - min problem that minimizes with respect to the structural design variables, the minimum value of the performance index with respect to the control forces obtained using the IMSC technique. The minimization process requires derivatives of eigenvalues and eigenvectors with respect to the design variables. These derivatives can be computed by a rather involved analytical procedure or a relatively simple finite difference procedure. The computer cost effectiveness of these two procedures for the derivative calculations is examined.

Description: The objective of this presentation is to introduce the attendees to the DYSCO program. The emphasis will be on the features which make it multidisciplinary. DYSCO is a very general and versatile software program which couples and solves dynamic systems. It was initiated in the late 1970's in response to a helicopter analysis requirement. The system development, however, resulted in an executive which was completely separated from any particular area of technology, except that of second order ordinary differential equations. During the course of its development, it was funded by the Army Aviation Applied Technology Directorate, the Air Force Wright Aeronautical Laboratories, and by the Kaman Aerospace Corporation. It is completely written in FORTRAN and is operational on IBM and VAX computers.

Description: A numerical method is presented for design sensitivity analysis, using an iterative-method reanalysis of the structure generated by a small perturbation in the design variable; a forward-difference scheme is then employed to obtain the approximate sensitivity. Algorithms are developed for displacement and stress sensitivity, as well as for eignevalues and eigenvector sensitivity, and the iterative schemes are modified so that the coefficient matrices are constant and therefore decomposed only once.

Description: The considerations and the resultant approach used to implement design sensitivity capability for grids into a large scale, general purpose finite element system (MSC/NASTRAN) are presented. The design variables are grid perturbations with a rather general linking capability. Moreover, shape and sizing variables may be linked together. The design is general enough to facilitate geometric modeling techniques for generating design variable linking schemes in an easy and straightforward manner. Test cases have been run and validated by comparison with the overall finite difference method. The linking of a design sensitivity capability for shape variables in MSC/NASTRAN with an optimizer would give designers a powerful, automated tool to carry out practical optimization design of real life, complicated structures.

Description: Structural optimization has been available to the structural analysis community as a tool for many years. The popular use of displacement method finite-element techniques to analyze linearly elastic structures has resulted in an ability to calculate the weight and constraint gradients inexpensively for numerical optimization of structures. Here, recent experiences in the investigation and use of structural optimization are discussed. In particular, experience with the commercially available ADS/NASOPT code is addressed. An overview of the ADS/NASOPT procedure and how it was implemented is given. Two example problems are also discussed.

Description: ASTROS (Automated Structural Optimization System) is a finite-element-based multidisciplinary structural optimization procedure developed under Air Force sponsorship to perform automated preliminary structural design. The design task is the determination of the structural sizes that provide an optimal structure while satisfying numerous constraints from many disciplines. In addition to its automated design features, ASTROS provides a general transient and frequency response capability, as well as a special feature to perform a transient analysis of a vehicle subjected to a nuclear blast. The motivation for the development of a single multidisciplinary design tool is that such a tool can provide improved structural designs in less time than is currently needed. The role of such a tool is even more apparent as modern materials come into widespread use. Balancing conflicting requirements for the structure's strength and stiffness while exploiting the benefits of material anisotropy is perhaps an impossible task without assistance from an automated design tool. Finally, the use of a single tool can bring the design task into better focus among design team members, thereby improving their insight into the overall task.

Description: Large space structures (LSSs) and other dynamical systems of current interest are often extremely complex assemblies of rigid and flexible bodies subjected to kinematical constraints. A formulation is presented for the governing equations of constrained multibody systems via the application of singular value decomposition (SVD). The resulting equations of motion are shown to be of minimum dimension.

Description: A recently developed finite-element capability for general nonlinear shell analysis, featuring the use of three-dimensional constitutive equations within an efficient resultant-oriented framework, is employed to simulate the postbuckling response of an axially compressed composite cylindrical panel with a circular cutout. The problem is a generic example of modern composite aircraft components for which postbuckling strength (i.e., fail-safety) is desired in the presence of local discontinuities such as holes and cracked stiffeners. While the computational software does a reasonable job of predicting both the buckling load and the qualitative aspects of postbuckling (compared both with experiment and another code) there are some discrepancies due to: (1) uncertainties in the nominal layer material properties, (2) structural sensitivity to initial imperfections, and (3) the neglect of dynamic and local material delamination effects in the numerical model. Corresponding refinements are suggested for the realistic continuation of this type of analysis.

Description: The development of commercial finite element software is addressed. This software provides practical tools that are used in an astonishingly wide range of engineering applications that include critical aspects of the safety evaluation of nuclear power plants or of heavily loaded offshore structures in the hostile environments of the North Sea or the Arctic, major design activities associated with the development of airframes for high strength and minimum weight, thermal analysis of electronic components, and the design of sports equipment. In the more advanced application areas, the effectiveness of the product depends critically on the quality of the mechanics and mechanics related algorithms that are implemented. Algorithmic robustness is of primary concern. Those methods that should be chosen will maximize reliability with minimal understanding on the part of the user. Computational efficiency is also important because there are always limited resources, and hence problems that are too time consuming or costly. Finally, some areas where research work will provide new methods and improvements is discussed.

Description: An analytical method is being developed to determine the signature of an acoustic emission waveform from a growing crack and the results of this analysis are compared to experimentally obtained values. Within the assumptions of linear elastic fracture mechanics, a two dimensional model is developed to examine a semi-infinite crack that, after propagating with a constant velocity, suddenly stops. The analytical model employs an integral equation method for the analysis of problems of dynamic fracture mechanics. The experimental procedure uses an interferometric apparatus that makes very localized absolute measurements with very high fidelity and without acoustically loading the specimen.

Description: The characteristics of the Synthetic Aperture Radar (SAR) antenna aboard the ESA Remote Sensing Satellite (ERS-1) are presented. The antenna is folded into a dense package for launch and is deployed in orbit. The design requirements and constraints, their impact on the design, and the resulting features of the mechanisms are discussed.

Description: A computer program was developed to take a NASTRAN output file from a normal modes analysis and calculate the modal strain energies of selected elements. The FORTRAN program can determine the modal strain energies for CROD, CBAR, CELAS, CTRMEM, CQDMEM2, and CSHEAR elements. Modal strain energies are useful in estimating damping in structures.

Description: Many finite element preprocessors describe finite element model geometry with points, lines, surfaces and volumes. One method for describing these basic geometric entities is by use of parametric cubics which are useful for representing complex shapes. The lines, surfaces and volumes may be discretized for follow on finite element analysis. The ability to limit or selectively recover results from the finite element model is extremely important to the analyst. Equally important is the ability to easily apply boundary conditions. Although graphical preprocessors have made these tasks easier, model complexity may not lend itself to easily identify a group of grid points desired for data recovery or application of constraints. A methodology is presented which makes use of the assignment of grid point locations in parametric coordinates. The parametric coordinates provide a convenient ordering of the grid point locations and a method for retrieving the grid point ID's from the parent geometry. The selected grid points may then be used for the generation of the appropriate set and constraint cards.