Real-time Optimal Energy Management System for Plug-in Hybrid Electric Vehicles
Loading...
Date
2014-09-24
Authors
Taghavipour, Amir
Journal Title
Journal ISSN
Volume Title
Publisher
University of Waterloo
Abstract
Air pollution and rising fuel costs are becoming increasingly important concerns for the transportation industry. Hybrid electric vehicles (HEVs) are seen as a solution to these problems as they off er lower emissions and better fuel economy compared to conventional internal combustion engine vehicles. A typical HEV powertrain consists of an internal combustion engine, an electric motor/generator, and a power storage device (usually a battery). Another type of HEV is the plug-in hybrid electric vehicle (PHEV), which is conceptually similar to the fully electric vehicle. The battery in a PHEV is designed to be fully charged using a conventional home electric plug or a charging station. As such, the vehicle can travel further in full-electric mode, which greatly improves the fuel economy of PHEVs compared to HEVs.
In this study, an optimal energy management system (EMS) for a PHEV is designed to minimize fuel consumption by considering engine emissions reduction. This is achieved by using the model predictive control (MPC) approach. MPC is an optimal model-based approach that can accommodate the many constraints involved in the design of EMSs, and is suitable for real-time implementations. The design and real-time implementation of such a control approach involves control-oriented modeling, controller design (including high-level and low-level controllers), and control scheme performance evaluation. All of
these issues will be addressed in this thesis.
A control-relevant parameter estimation (CRPE) approach is used to make the control-oriented model more accurate. This improves the EMS performance, while maintaining its real-time implementation capability.
To reduce the computational complexity, the standard MPC controller is replaced by its explicit form. The explicit model predictive controller (eMPC) achieves the same performance as the implicit MPC, but requires less computational effort, which leads to a fast and reliable implementation. The performance of the control scheme is evaluated through different stages of model-in-the-loop (MIL) simulations with an equation-based and validated high-fidelity simulation model of a PHEV powertrain.
Finally, the CRPE-eMPC EMS is validated through a hardware-in-the-loop (HIL) test. HIL simulation shows that the proposed EMS can be implemented to a commercial control hardware in real time and results in promising fuel economy figures and emissions performance, while maintaining vehicle drivability.
Description
Keywords
Plug-in hybrid electric vehicles, Energy management system, Fuel economy, Emissions, Explicit model predictive control, Control-relevant parameter estimation, Hardware-in-the-loop simulation