Powerpack Optimal Design Methodology with Embedded Configuration Benchmarking 2016-01-0313
Design of military vehicle needs to meet often conflicting requirements such as high mobility, excellent fuel efficiency and survivability, with acceptable cost. In order to reduce the development cost, time and associated risk, as many of the design questions as possible need to be addressed with advanced simulation tools. This paper describes a methodology to design a fuel efficient powerpack unit for a series hybrid electric military vehicle, with emphasis on the e-machine design. The proposed methodology builds on previously published Finite element based analysis to capture basic design features of the generator with three variables, and couples it with a model reduction technique to rapidly re-design the generator with desired fidelity. The generator is mated to an off the shelf engine to form a powerpack, which is subsequently evaluated over a representative military drive cycles. An iterative procedure is developed, in which the optimization of the supervisory controller is embedded into the design optimization framework. Therefore, for every combination of design parameters the Dynamic Programming routine develops a benchmark control for minimum fuel consumption. This ensures realistic numbers for every function call, and convergence on a true optimum. Results can then guide the development of the new generator for a selected production engine and SHEV configuration.