Showing papers by "Thomas K. Caughey published in 1992"

Structure-unknown non-linear dynamic systems: identification through neural networks

Abstract: Explores the potential of using parallel distributed processing (neural network) approaches to identify the internal forces of structure-unknown non-linear dynamic systems typically encountered in the field of applied mechanics. The relevant characteristics of neural networks, such as the processing elements, network topology, and learning algorithms, are discussed in the context of system identification. The analogy of the neural network procedure to a qualitatively similar non-parametric identification approach, which was previously developed by the authors for handling arbitrary non-linear systems, is discussed. The utility of the neural network approach is demonstrated by application to several illustrative problems.

A Theorem on the Exact Nonsimilar Steady-State Motions of a Nonlinear Oscillator

Abstract: In this work the steady-state motions of a nonlinear, discrete, undamped oscillator are examined. This is achieved by using the notion of exact steady state, i.e., a motion where all coordinates of the system oscillate equiperiodically, with a period equal to that of the excitation. Special forcing functions that are periodic but not necessarily harmonic are applied to the system, and its steady response is approximately computed by an asymptotic methodology. For a system with cubic nonlinearity, a general theorem is given on the necessary and sufficient conditions that a excitation should satisfy in order to lead to an exact steady motion. As a result of this theorem, a whole class of admissible periodic functions capable of producing steady motions is identified (in contrast to the linear case, where the only excitation leading to a steady-state motion is the harmonic one). An analytic expression for the modal curve describing the steady motion of the system in the configuration space is derived and numerical simulations of the steady-state motions of a strongly nonlinear oscillator excited by two different forcing functions are presented.

An Experimental Study of the Active Control of a Building Model

Abstract: The active control of large structural systems is a subject of growing worldwide interest. One of the reasons for the increasing attention is the successful application of passive structural control methods such as base-isolation approaches and damping augmentation techniques. Research activity in the civil engineering field has been primarily focused on theoretical studies with few, limited experimental investigations. This paper reports some of the results of an ongoing analytical and experimental study into the control of building-like structures subjected to nonstationary random excitations such as earthquakes. The structural model used resembles a 5-story building about 2.5 meters high. The building model was subjected to a variety of direct-force excitations. The control algorithm used employes an adaptive structural member at a pre-determined location in the model in order to attenuate the structural response relative to the moving building foundation. An electromagnetic actuator is used to generate the required control forces in the "smart" member. Among the key features of the algorithm under discussion are: 1. Only one active controller is required to attenuate the vibration response contributed by the first three modes; the damping factor is increased from virtually zero to about 20%. 2. Only two sensors are needed for this algorithm; this leads to simpler instrumentation and a more robust system. 3. Due to the optimization procedure used to select the controller location, a significant amount of damping augmentation is obtained from a relatively small amount of control energy. 4. The whole design procedure was demonstrated, especially attention was paid to time lag problem of the active controller and the stability of the system was discussed. As part of the design phase of this study, a system identification procedure was used to develop a suitable reduced-order mathematical model. The results of a simulation study using this identified model are compared to experimental measurements. Problems encountered in the experimental phase of the study are reported and discussed. It is shown that (1) the algorithm under discussion is capable of reliably controlling the motion of the test structure under arbitrary dynamic environments, and (2) the features of the algorithm makes it a promising candidate for application to large civil structures.

The Effect of Inlet Swirl on the Rotordynamic Shroud Forces in a Centrifugal Pump

Abstract: The role played by fluid forces in determining the rotordynamic stability of a centrifugal pump is gaining increasing attention. The present research investigates the contributions to the rotordynamic forces from the discharge-to-suction leakage flows between the front shroud of the rotating impeller and the stationary pump casing. In particular, the dependency of the rotordynamic characteristics of leakage flows on the swirl at the inlet to the leakage path was examined . An inlet guide vane was designed for the experiment so that swirl could be introduced at the leakage flow inlet. The data demonstrates substantial rotordynamic effects and a destabilizing tangential force for small positive whirl ratios; this force decreased with increasing flow rate. The effect of swirl on the rotordynamic forces was found to be destabilizing.