EAGER: The Surrogate System Hypothesis for Joint Mechanics

William Marsh Rice University, Houston TX

Investigators

Abstract

This project is aimed at understanding the fundamental physics governing the behavior of mechanical joints in engineered structures. Mechanical joints are essential and integral parts of engineered systems and yet the physics governing them is largely unknown. This lack of knowledge prevents engineers and designers from developing predictive models of a jointed structure that can be used to guide its design. As a result, jointed structures are often overdesigned to prevent failure from occurring. In low consequence applications, such as an office chair, home electronic appliance, etc., overdesign of the jointed interface has little practical consequence. For high consequence applications, such as aero-turbines for passenger jets, the fuselage of rockets, defense applications, or the frame of an automobile, overdesign can have significant ramifications in terms of fuel efficiency and safety. If a predictive model of jointed structures existed (i.e. one that predicts both the stiffness of a joint and how much energy it dissipates), then joints could be properly designed contributing their appropriate weight to the structure, with the potential of reducing fuel consumption. A second outcome is that the predictive model could be used in design of jointed interfaces to dissipate energies that might otherwise damage other components of the assembly, such as sensitive electronics within a rocket. This EArly-concept Grant for Exploratory Research (EAGER) project seeks to address this gap in predictive capability by developing a rigorous framework, the surrogate system hypothesis, in which a jointed interface can be characterized out of context (avoiding complications associated with high manufacturing expenses for the real system or multiple sources of contamination in measurements such as from other joints within the system) and then used to predict the response of the joint within the system of interest with high accuracy. The surrogate system hypothesis states that a surrogate structure that contains the same joint as the system of interest can be used to deduce the properties of the joint. These properties, once accounting for the properties of the surrogate structure, can then be substituted directly into the system of interest as a spatially discrete joint model (as opposed to a modal model). The goal of this project is to test the surrogate system hypothesis to determine if, under varied loading conditions and in realistic structures, it is supported by experimental evidence. The experimental investigations must, by nature of the hypothesis, be in combination with numerical modeling efforts, which will be built upon recently matured reduced order modeling techniques from the dynamic substructuring community for studying the response of a jointed structure. The work is divided into several objectives: testing the surrogate system hypothesis in new regimes in order to challenge its underlying assumptions, determining the bounds of the hypothesis' limitations, and demonstrating proof-of-concept on a realistic system with more complex geometries. If the surrogate system hypothesis finds support, it will enable several significant advances for contact mechanics in addition to the design and optimization of assemblies. This research will enable a new approach for predictive models of joints based off of well-characterized surrogate systems. Once a predictive approach is developed, it will be possible to detect potential failures at the design phase before hardware is manufactured. The ramification of this research is that, if successful, this hypothesis indicates that micro- and nano-scale effects, such as the distribution of asperities, are secondary to determining the dynamics of a structure behind the macro-scale features such as geometry and mean roughness.