Search Results

Polyurethane foam is often a major constituent of automotive seating, and exhibits highly nonlinear behavior under normal operating conditions. Efficient design requires an understanding, as well a good model, of the foam behavior. The work presented here is an attempt to link continuum and microstructural approaches to modeling foamed materials and take advantage of the utility in each. The outcome will ultimately be the ability to generate a foam superelement that is sensitive to microstructural properties but does not require the computational complexity of a microstructural finite element model. This will facilitate the iterative design of seating for comfort and other dynamic considerations. To this end, an Ogden-type continuum model for compressible rubber-like solids, is fitted to the results of numerous simulated compression tests conducted on finite element models of two-dimensional foam.

Hip joint location (H-point) is an important design specification used by car seat manufacturers. Since most of the modern car seats are full-foam seats, the H-point location is primarily dependent on the quasi-static behavior of foam which is a highly nonlinear and viscoelastic material. In this work, a constitutive model is developed for flexible polyurethane foam. The stiffness characteristics of foam are described by a polynomial in displacement and the viscoelastic behavior is described by a convolution of the response with a kernel which is assumed to be a sum of exponentials. A system identification procedure, based on linear least squares fitting and ARMA modeling, is developed to identify the parameters from data collected in quasi-static foam experiments.

An overview of model development for seated occupants is presented. Two approaches have been investigated for modeling the vertical response of a seated dummy: finite element and simplified mass-spring-damper methods. The construction and implementation of these models are described, and the various successes and drawbacks of each modeling approach are discussed. To evaluate the performance of the models, emphasis was also placed on producing accurate, repeatable measurements of the static and dynamic characteristics of a seated dummy.