The research objective of this award is the development of a crashworthy hybrid cellular automata (crasHCA) method for nonlinear-dynamic transient topology and topography optimization. Crashworthiness design methods are used to synthesize vehicle structures to protect occupants against injury. Currently, the HCA approach for structural design is limited to linear static analysis for structural synthesis. In addition, competing topology optimization approaches require sensitivity calculations and are therefore limited to linear static analysis. Because analytic expressions for sensitivities or numerical approaches for sensitivity calculation in nonlinear transient problems are not available, current practice is limited to the use of surrogate models for design optimization. The results of this research will provide users with a convergent nonlinear transient topology and topography optimization approach for crashworthiness design. The HCA framework to date has been limited by the ability to use only brick elements for topology optimization. In this investigation the HCA methodology will include the use of shell elements within a framework referred to as topography optimization for the design of stamped sheet metal components.
If successful this proposal will facilitate novel, new designs that will lead to increased levels of protection for both drivers and passengers in automotive vehicles. In the automotive industry the use of topology optimization for structural component design is increasingly popular. The development of a topology optimization capability that facilitates nonlinear transient topology and topography optimization (crasHCA) will allow vehicle designers to significantly improve vehicle safety in reduced product development cycle times. CrasHCA topology optimization will allow designers to move beyond surrogate models and to facilitate structural synthesis under nonlinear transient dynamic loading. In this GOALI investigation the PI and students will interact with Honda R&D Americas to insure technology transfer to industry. The crasHCA methodology for nonlinear transient topology optimization will be validated using real application test problems provided by Honda. Fabrication and mechanical testing of the synthesized designs at Honda?s research facility will further contribute to verification and validation.
This industry-university collaborative project combines the product development expertise of Honda R&D Americas, Inc. with the advanced numerical optimization algorithm development expertise of the University of Notre Dame. Honda R&D Americas is committed to "Safety for Everyone" through technology and innovation. If successful this proposal will facilitate design tool developments that will lead to increased levels of protection for both drivers and passengers in automotive vehicles. The main outcome of this investigation is a novel crashworthy hybrid cellular automata (crasHCA) method for nonlinear transient topology and topography optimization. The original HCA method, developed by Notre Dame researchers, provides a linear static topology optimization capability. The HCA optimization tool is a gradient free approach that was originally developed for simulating the bone remodeling process. The resulting HCA method has been used for topology optimization facilitating structural synthesis and compliant mechanism design using linear finite element analysis. The initial development efforts were supported by a Honda Initiation Grant. The PI and students have worked closely with researchers at Honda R&D Americas during the HCA development effort. In this investigation the HCA methodology will be extended to include fully nonlinear transient finite element analysis for crashworthiness topology optimization (crasHCA). Note that competing topology optimization approaches require sensitivity calculations and are therefore limited to linear static analysis. Analytic expressions for sensitivities or numerical approaches for sensitivity calculation in nonlinear transient problems are not available. Development of the crasHCA tool will provide users with a convergent nonlinear transient topology and topography. The HCA framework has been extended into a comprehensive tool for crashworthy structural optimization referred to as crasHCA. The algorithm applied to topology optimization has been incorporated by LSTC as LS-TaSC in LS-DYNA environment. The HCA has been further extended to topography (thickness) optimization. Main accomplishments include the development of design approaches for controlled energy absorption (i.e., prescribed force-displacement behavior). The two most promising approaches are the use sub-domains (i.e., flexible and stiff) for continuum structures and the use of compliant mechanisms in tubes (i.e., tubular compliant mechanisms). In particular, the use of compliant tubular structures allows the generation of robust structures insensitive to variations in the loading condition and geometric imperfections. In the automotive industry the use of topology optimization for structural synthesis is increasingly popular. Designers are currently limited to linear finite element based topology optimization applications. The development of a HCA tool that facilitates nonlinear transient topology and topography optimization (crasHCA) will allow vehicle designers to significantly improve vehicle safety in reduced product development cycle times. Current practice in crashworthiness design is limited to the use of surrogate models for nonlinear transient design optimization. The PI and students will continue to interact with Honda R&D Americas to insure technology transfer to industry. The crasHCA methodology for nonlinear transient topology optimization will be validated using real application test problems provided by Honda. This grant supported: • Four Ph.D. students: Chandan Mozumder (graduated in 2010), Punit Bandi (expected graduation in 2012), John Goetz (expected graduation in 2013), and Andrew Baumann (expected graduation in 2014) • One post-doc: Palani Ramu (2009) • Four undergraduate students in a 10-week research experience: Malloy, Dennis, 2009; Dudley, Brittany in 2009; Kinnary, Kyle, 2010; Henisey, Teresa 2010. The results have been published in: • 20 conference proceedings • Seven journal articles • One chapter in book. The educational impact includes: • One One K-12 teacher has been supported by this grant: Mr. Michael Lewis (2008) • Development of a course on Topology Optimization that is now part of the Aerospace and Mechanical Engineering curricula at the University of Notre Dame.
