Engineering topics
Publications and source records attributed to Kuo, C. P..
Testing under variety of support conditions combined with computer analysis to update mathematical models to match test data. Report suggests dynamic characteristics of large space structures predicted, without full-scale testing, by method that combines experiment and analysis. Method, multiple-boundary-condition testing, developed for such large space structures as dish antennas, towers, and solar-cell arrays.
Explore the source record for details and available documents.
The Multiple Boundary Condition Tests (MBCT) approach is evaluated and recent modifications are described. For the application of MBCT, 12 different boundary conditions are selected and the results of applying MBCT in conjunction with a nonlinear formulation are indicated schematically. It is concluded that the nonlinear formulation enhances the ability to implement the MBCT test approach on large space structures which cannot be ground tested without the artificial boundary conditions incorporated in MBCT. In addition, it leads to significant improvements in the convergence to the correct solution.
The evaluation of sensitivity coefficients which are the changes in the eigencharacteristics with respect to changes in the structural parameters is important in the identification of structural systems. In such studies the commonly used linearized sensitivity coefficients for system identification have occasionally resulted in poor convergence or even divergence during the iteration process to update analytical models for better correlation with the test data. Nonlinear sensitivity coefficients and nonlinear correction terms, usually eliminated during the linearization process, have been developed to evaluate sensitivity coefficients of linear systems. Application to computer simulated example problems indicates a more rapid and stable convergence of the iteration processes to update the analytical model for the improvement of the correlation with test data. The successful application of combining the nonlinear sensitivity coefficients and nonlinear correction terms with the Multiple Boundary Condition Tests enhances the ability to validate large space structures by ground tests.
The paper presents concepts designed to validate, through the use of ground tests, mathematical models of continuous type structures and structures comprised of interconnecting subsystems. For continuous-type structures, a multiple boundary condition test approach is considered in which the basic idea is to perform a large number of tests using artificial boundary conditions from which good ground test data can be obtained. The test ensures an arbitrarily large number of test data for use in the validation and updating of the mathematical model. Another approach, applicable to structures comprised of subsystems, involves the identification of significant structural elements for the system dynamic model which are not validated by standard modal tests of the subsystems.
A major concern for future large space structures is the ability to verify their dynamic characteristics by ground test. This article presents the results of the multiple boundary conditions test (MBCT) approach, which provides a complete ground test of a large structure that will provide, in turn, the data necessary to construct a test-verified final mathematical model. Theoretical studies indicate that this approach can provide a better final model than a ground test of the full-scale very flexible structure in a 1-g field. The approach is demonstrated by a mathematical simulation.
The finite element method is used to investigate stress/strain in silicon ribbon. Failure considerations such as residual stress, buckling material non-linearity and creep are discussed. Temperature profiles are presented.
An advanced leading-edge concept was analyzed using the space shuttle leading edge system as a reference model. The comparison indicates that a direct-bond system utilizing a high temperature (2700 F) fibrous refractory composite insulation tile bonded to a high temperature (PI/graphite) composite structure can result in a weight savings of up to 800 lb. The concern that tile damage or loss during ascent would result in adverse entry aerodynamics if a leading edge tile system were used is addressed. It was found from experiment that missing tiles (as many as 22) on the leading edge would not significantly affect the basic force-and-moment aerodynamic coefficients. Additionally, this concept affords a degree of redundancy to a thermal protection system in that the base structure (being a composite material) ablates and neither melts nor burns through when subjected to entry heating in the event tiles are actually lost or damaged during ascent.
A nonparametric identification technique for the identification of close coupled dynamic systems with arbitrary memoryless nonlinearities is presented. The method utilizes noisy recorded data (acceleration, velocity and displacement) to identify the restoring forces in the system. The masses in the system are assumed to be known (or fairly well estimated from the design drawings). The restoring forces are expanded in a series of orthogonal polnomials and the coefficients of these polynomial expansions are obtained by using least square fit method. A particularly simple and computationally efficient method is proposed for dealing with separable restoring forces. The identified results are found to be relatively insensitive to measurement noise. An analysis of the effects of measurement noise on the quality of the estimates is given. The computations are shown to be relatively quick (when compared say to the Wiener identification method) and the core storage required relatively small, making the method suitable for onboard identification of large space structures.
A relatively simple nonparametric method for the identification of a class of close-coupled nonlinear multi-degree-of-freedom systems has been developed. The identification of arbitrary memoryless nonlinearities is possible through knowledge of the accelerations, velocities, and displacements of the various masses. These quantities are used to obtain the surfaces of the restoring forces as functions of the intermass displacements and velocities. The method was demonstrated by application to a four-degree-of-freedom system to identify the restoring forces. It is found that the identification results are relatively insensitive to measurement noise.