In the automotive and aerospace industries, every component experiences dynamic loading. The complex, multi-axial loads and surrounding system are the environmental conditions. Due to the lack of understanding of nonlinear system dynamics, the components and structures have been over-designed to handle these environmental conditions and reduce nonlinear influences. Over-designing leads to an increase in weight, which in turn reduces fuel efficiency.

In recent years, there has been a drive to improve fuel economy resulting in the redesign of these over-designed systems. Reducing the component conservativeness increases nonlinear influences on the dynamic response of the system. Research is ongoing to characterize these new systems in laboratory settings. However, there is a disconnect between the environmental conditions and laboratory experiments. These disconnects are a result of improper model simplifications and input forces, costing billions of dollars in inappropriate/ additional qualification tests. Currently, researchers are using varying levels of fidelity to model the salient physics of the nonlinear systems. Unfortunately, they are still unable to match the response of systems with strong nonlinearities fully.

This thesis is motivated by the need to improve and develop experimental methods to more accurately update models for environmental conditions. The specific issues addressed and significant contributions include:

1. A refinement of a recently formalized method proposed by D.J. Ewins, which consists of ten steps to perform model validation of nonlinear structures. This work details through a series of experimental studies that many standard test setup assumptions and properties of nonlinearities are invalid. This invalidation is due to both known and unrecognized nonlinear properties. A review of current methods for characterizing nonlinear and gaps in the approaches.
2. A series of tests to determine the influence of multi-axial excitation on system responses. Due to the complex environmental loads, a component will not experience excitation from only one direction. Current laboratory tests qualify the system one axis at a time, under the assumption that a nonlinearity has no dependence on the load direction.
3. A design methodology to create a test fixture to emulate the environmental boundary condition. The methodology truncates boundaries to a more reasonable size for laboratory testing by utilizing spring-mass systems to account for the loss of mass and stiffness. The method is validated by comparing stress states of an attached component of the full system and truncated one.