ORNL, UT Austin team proposes optimization framework for hybrids; balancing fuel consumption, motor efficiency, battery capacity and life
Researchers at Oak Ridge National Laboratory and the University of Texas at Austin have developed an optimization framework for hybrid-electric and plug-in hybrid-electric vehicles (HEVs and PHEVs) including fuel consumption, motor efficiency, and battery capacity and lifetime. The approach, detailed in a paper accepted for publication in the IEEE Transactions On Control Systems Technology, is intended to enhance the understanding of the associated tradeoffs among the HEV subsystems—e.g., engine, motor, battery—and to investigate the related implications for fuel consumption and battery capacity and lifetime.
With the framework, the performance of the subsystems can be tailored according to consumer preferences, such as reducing fuel consumption or extending battery life. Further, the ability to control battery performance indices—e.g., temperature—enables operating the battery at a higher target SoC (state of charge) without incurring safety concerns. As a result, the results of such analyses could have significant implications for the related HEV and PHEV ownership and warranty costs.
The main advantage of HEVs and PHEVs is the existence of individual subsystems, e.g., the internal combustion engine, the electric machines (motor and generator), and the energy storage system (battery) that can power the vehicle either separately or in combination. The supervisory power manage-ment control algorithm determines how to distribute the power demanded by the driver to these subsystems.
Traditionally, maximizing fuel economy and minimizing GHG emissions are the main objectives of the control development. With the introduction of batteries, prolonging the battery life is considered highly important during the control design from the consumer’s perspective due to associated ownership and warranty costs. Thus, it is necessary to develop a control framework for vehicle power management that will be able to balance maximum fuel economy, minimum GHG emissions, and extended battery life based on consumer’s needs and preferences.
… The objective of this paper and related research of the authors is to enhance our understanding of the associated tradeoffs among the HEV subsystems, e.g., the engine, the motor, and the battery. Addressing these tradeoffs and related implications in fuel consumption, motor efficiency, and battery capacity and lifetime can aid in developing power management control algorithms that prioritize these objectives based on consumer needs and preferences.—Shaltout et al.
In this work, the researchers developed a multiobjective optimization framework that includes fuel consumption; motor efficiency; and the charge/discharge process to indirectly address battery use and lifetime. They also implemented a power management control algorithm yielding the Pareto optimal solution of the multiobjective optimization problem that minimizes the long-run expected average cost criterion in the stochastic control problem formulation.
The vehicle model was a pre-transmission parallel HEV model, with the hybrid propulsion system consisting of a gasoline engine coupled to an automatic transmission through a gear set and a clutch. The electric machine (motor/generator) is coupled through another gear set to the transmission input shaft. Both gear sets have been assigned to have unity gear ratios. The transmission output shaft is coupled to a final drive. The electric path of the hybrid propulsion system consists of the electric machine connected to a rechargeable battery. The available control variables are the engine and motor torques as the engine and motor speeds are determined by the vehicle speed.
They modeled the different components of the hybrid propulsion system—except for the battery—in Autonomie. Autonomie is a MATLAB/Simulink simulation package for powertrain and vehicle model development developed by Argonne National Laboratory. They replaced the battery model in Autonomie with their own.
The new computationally efficient battery model is able to capture charge–discharge behavior well to study thermal response of the batteries. Although for the paper they worked with a NiMH battery system, the analysis can be easily extended to a Li-ion battery system. The battery model is derived from the experimental discharge characteristics of the cell.
They used thee driving cycles in their analysis: the urban dynamometer driving schedule (UDDS); the FTP; and the New York. They compared the Pareto control policy corresponding to scenarios with and without the battery, to investigate the associated tradeoffs.
|Tradeoff analysis between the battery performance and fuel consumption at 60% and 70% target SoC. Shaltout et al. Click to enlarge.|
Ultimately, the proposed optimization framework establishes the foundation for addressing overall HEV optimization, including additional performance characteristics. Thus, paving many pathways for future research work. For instance, reducing the engine emissions can be considered in the multiobjective cost function, thus adding another element to the aforementioned set of consumer preferences. In addition, the proposed framework allows for augmenting the control inputs to include the transmission gearshift schedule as a separate control input, thus enhances the HEV optimal control problem.—Shaltout et al.
This work was supported in part by the Laboratory Directed Research and Development Program through the Oak Ridge National Laboratory.
Mohamed L. Shaltout, Andreas A. Malikopoulos, Sreekanth Pannala, and Dongmei Chen, “A Consumer-Oriented Control Framework for Performance Analysis in Hybrid Electric Vehicles” IEEE Transactions On Control Systems Technology doi: 10.1109/TCST.2014.2376472