Optimal power control methodology with its application to plug-in hybrid electric vehicles

Abstract

The depletion of fossil oil reserves and the growing concerns on the effects of global warming have increased the popularity of plug-in hybrid electric vehicles (PHEVs). In this study, the power control problem for PHEVs, including power management and vehicle charging control problems, is investigated from an optimal control perspective. Two novel power management strategies are developed for microturbine (MT)-powered and proton exchange membrane fuel cell (PEMFC) PHEVs, which optimally control the power flow of the powertrain at the vehicle operational mode. An optimal vehicle charging strategy is proposed to regulate the vehicle charging rate from the transformer, which can shift the transformer load during the vehicle charging process. This thesis covers the following aspects: First, a series hybrid midsize sedan with an MT and a chargeable Li-ion battery stack as its primary power source and energy storage system, respectively, is modeled. A novel approach called telemetry equivalent consumption minimization strategy (T-ECMS) is proposed to minimize driving cost according to Pontryagin’s minimum principle (PMP). This approach solely relies on the vehicle position detected with a telemetry system and the measured battery state of charge (SOC). Simulation results show that the T-ECMS approach exhibits control performance equivalent to that determined from deterministic dynamic programming in terms of driving cost and diesel consumption. The proposed approach significantly reduces the driving cost from 7.7% to 21.6% over both urban and highway cycles compared with a baseline control. Given that this strategy uses feedback from the battery SOC, control performance is insensitive to control parameter errors. Second, the power management control problem for a series plug-in PEMFC/Li-ion battery hybrid midsize sedan is formulated and investigated using a two-stage controller (TSC) that minimizes hydrogen consumption and protects the PEMFC lifetime. The proposed TSC consists of two controllers designed with different control functions in two stages. During the first stage of design, a predictive controller is developed based on the T-ECMS approach to predict the global battery SOC optimality trend and local control reference without considering PEMFC lifetime constraints. During the second stage of design, a tracking controller is designed to track the local control reference with respect to PEMFC lifetime constraints and other physical limitations at the current control step, thus ensuring that the system follows the optimal battery SOC reference over a long time horizon. Finally, simulation results show that the TSC achieves a reasonable trade-off between hydrogen consumption and PEMFC lifetime protection. Third, the optimal vehicle charging control problem for PHEVs with bidirectional power flow is formulated and investigated at the residential transformer level. A twostage charging control (TSCC) strategy is proposed to shift the transformer load while achieving good charging performance for all PHEVs connected to the grid. The proposed TSCC consists of an aggregator optimizer and a power distributor designed with different control functions in two stages. During the first stage, the optimal charging power of all PHEVs in the aggregator is derived from the PMP based on the concept of dynamic aggregator. During the second stage, a power distribution law is developed to allocate the aggregated power from the first stage by using fuzzy logic control. The TSCC approach considers the stochastic characterization of practical vehicle charging scenarios and can therefore be implemented in real time. Finally, simulation results are presented to validate the control performance of the TSCC. In summary, three optimal power control methods (i.e., T-ECMS, TSC, and TSCC) are developed to address the power management and vehicle charging control problems for PHEVs. For the power management control problem, the T-ECMS is designed to control the power flow of the MT PHEV in real time, which can achieve the least driving cost with small computational time. The TSC proposed for the PEMFC PHEV can result in minimum hydrogen consumption and effective PEMFC lifetime protection by considering PEMFC lifetime constraints. For the vehicle charging control problem, the TSCC can significantly reduce the transformer load peak and fully charge all PHEVs at the end of the charging process.

Date of Award	15 Jul 2013
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Dong SUN (Supervisor)