Chassis weight reduction in military vehicles and the advantages of utilizing High Performance Computing

Military vehicles including the HMMWV (High Mobility Multipurpose Wheeled Vehicle, also called the Humvee) have been required to add anti-ballistic armor to shield against IED's (Improvised Explosive Devices) and other ballistics. This weight addition was not included in the vehicles' original design and as a result many vehicles are fatiguing much quicker and breaking down in fields of combat. In order to combat this weight increase, our team has investigated replacing steel components in the vehicle chassis with advanced composite/steel hybrid designs. Baseline stiffness and strength analysis were first established with current designs/materials. The hybrid design optimizations were conducted utilizing the Genesis software package from Vanderplaats R & D. Weight reductions of 30% were achieved with same stiffness hybrid designs of a subassembly (reductions up to 65% resulted on a per part basis). To analyze the static strength and fatigue life Alpha Star's Genoa software package was used. Static strength increased by 20% while fatigue life increased by 75%. The advantages of utilizing HPC (High-Performance Computing) for these analyses was also investigated. In the past, vehicle component redesigns and optimizations have been performed on a system or subsystem level. General assumptions of the boundary conditions on the outer envelope of the systems have to be made in order to obtain convergent solutions. These assumptions could be by-passed by performing whole vehicle modeling. Working with the OSC (Ohio Supercomputer Center) our team measured calculation time for static durability analyses of various models of complexity and number of elements. It was determined that for small to medium sized models, which would include component and subsystem analyses, the advantage of using HPC was minimal. However, for much larger models, which would include entire vehicles, a 38% reduction in run time was achieved compared to what could be modeled with state of the art workstations. This advantage was even greater for complex fatigue modeling, resulting in a run time reduction of 52%.