Model-Based Design of a Hybrid Powertrain Architecture with Connected and Automated Technologies for Fuel Economy Improvements

Simulation-based design of connected and automated hybrid-electric vehicles is a challenging problem. The design space is large, the systems are complex, and the influence of connected and autonomous technology on the process is a new area of research. The Ohio State University EcoCAR Mobility Challenge team developed a comprehensive design and simulation approach as a solution. This paper covers the detailed simulation work conducted after initial design space reduction was performed to arrive at a P0-P4 hybrid vehicle with a gasoline engine. Two simulation environments were deployed in this strategy, each with unique advantages. The first was Autonomie, which is a commercial software tool that is well-validated through peer-reviewed studies. This allowed the team to evaluate a wide range of components in a robust simulation framework. To ensure consistent evaluation between potential architectures, the team paired Autonomie with a particle swarm optimizer to automatically calibrate the hybrid supervisory control and achieve near optimal control calibrations. The team also utilized a dynamic programming model environment to evaluate the fuel economy impact of connected and autonomous systems onboard the vehicle. Dynamic programming provides a true optimum fuel economy solution using a complete understanding of the vehicle’s drive cycle and component performance. This approximates the upper end fuel economy, or what is achievable with complete knowledge of the drive trace. The team validated the simulation approaches using a variety of techniques, such as validating the ability to model the stock vehicle and demonstrating correlation of design to similar hybrid vehicles in production. Overall, the team presents a complete synopsis of a well-executed process used to arrive at a near optimal vehicle design for competing in the EcoCAR Mobility Challenge.