Real-Time Optimal Control of a Plug-in Hybrid Electric Vehicle Using Trip Information

Abstract

The plug-in hybrid electric vehicle (PHEV) is a promising option for future sustainable transportation. It offers better fuel economy and lower emissions than conventional vehicles. This thesis has developed a novel energy-optimal powertrain controller for PHEVs. The controller will be broadly applicable to all PHEV models; however, it will be fine-tuned to the Toyota Prius Plug-in Hybrid for testing and validation. The controller will take advantage of advancements in vehicle intelligent and communications technologies, such as Global Positioning System (GPS), Intelligent Transportation System (ITS), Geographic Information System (GIS), radar, and other on-board sensors, to provide look-ahead trip data. These data are critical to increasing fuel economy as well as driving safety. This PhD research has developed three energy-optimal systems for PHEVs: Trip Planning module, Route-based Energy Management System (Route-based EMS), and Ecological Cruise (Eco-Cruise) Controller. The main objective of these energy-optimal systems is to minimize the total energy cost, including both electricity derived from the grid and fuel. The upper-level system is Trip Planning, using an algorithm designed to take advantage of previewed trip information to optimize State of Charge (SOC) profiles. The Route-based EMS optimally distributes propulsion power between the batteries and engine. Finally, the Eco-Cruise controller adjusts the speed considering upcoming trip data. Real-time implementation has remained a major challenge in the design of complex control systems. To address this hurdle, simple and efficient models and fast optimization algorithms are developed for each energy-optimal strategy. A Real-time Cluster-based Optimization is developed to solve the Trip Planning problem in real-time. The Route-based EMS is developed based on Equivalent Consumption Minimization Strategy (ECMS) to optimally distribute propulsion power between two energy sources. And, a Nonlinear Model Predictive Control (NMPC) is utilized to obtain optimum traction or regenerative torques in Eco-Cruise controller. Model-in-the-Loop (MIL) and Hardware-in-the-Loop (HIL) testing are critical steps in control validation and in ensuring real-time implementation capability. The MIL results show that the novel energy-optimal powertrain controller can improve the total energy cost by up to %20 compare to benchmark rule-based controller. The HIL test results demonstrate that the computational time for energy-optimal strategies are less than the target sampling-time, and they can be implemented in real-time.