Modeling, configuration and control optimization of power-split hybrid vehicles

Hybrid electric vehicles (HEV) represent one of the most promising fuel-saving technologies in the short-term for improving fuel economy of ground vehicles. Their viability has been amply demonstrated in a few successful commercial models. Among the common configurations, the power-split (i.e., combined parallel/series) configuration offers superior design and control flexibility and achieves highest overall efficiency. Therefore, most of the full-hybrid vehicles planned for the near future from Toyota, Lexus, GM, Chrysler, and Ford are all split hybrids. In this dissertation, a model-based configuration and control optimization analysis of power-split HEV is presented.; An integrated dynamic model was first developed for power-split HEV powertrain systems. From this simulation model, a math-based universal model format is generalized. It presents different designs of power-split powertrains regardless of the various connections between gear nodes and power sources. Based on this universal format, a methodology that automatically generates dynamic models is developed. It not only enables rapid generation of powertrain models, but also allows the process of automatically exploring possible configuration designs.; We next introduce a design screening process and a combined configuration and control optimization strategy. In the design screening process, various design requirements including transmission efficiency, drivability, power source component sizing are utilized to evaluate possible configurations and select valid design candidates. In the combined configuration and control optimization strategy, a control design procedure based on deterministic dynamic programming (DDP) was employed to find the optimal operation of the vehicle system and achieve the performance benchmarks for different configuration candidates. The optimal design solution is then achieved by comparing these benchmarks. This methodology allows design engineers to study powertrain configurations more scientifically and efficiently.; Finally, with the DDP suggesting the potential performance benchmark of the selected powertrain configuration, two alternative control strategies, stochastic dynamic programming and equivalent consumption minimization strategy, are developed to approach this performance benchmark. Both of these two control designs can be implemented in real-time and show close agreement with the DDP results in the simulation.