Xuewei Qi

Data-Driven Reinforcement Learning–Based Real-Time Energy Management System for Plug-In Hybrid Electric Vehicles

Plug-in hybrid electric vehicles (PHEVs) show great promise in reducing transportation-related fossil fuel consumption and greenhouse gas emissions. Designing an efficient energy management system (EMS) for PHEVs to achieve better fuel economy has been an active research topic for decades. Most of the advanced systems rely either on a priori knowledge of future driving conditions to achieve the optimal but not real-time solution (e.g., using a dynamic programming strategy) or on only current driving situations to achieve a real-time but nonoptimal solution (e.g., rule-based strategy). This paper proposes a reinforcement learning–based real-time EMS for PHEVs to address the trade-off between real-time performance and optimal energy savings. The proposed model can optimize the power-split control in real time while learning the optimal decisions from historical driving cycles. A case study on a real-world commute trip shows that about a 12% fuel saving can be achieved without considering charging opportunities; further, an 8% fuel saving can be achieved when charging opportunities are considered, compared with the standard binary mode control strategy.

Development and Evaluation of an Evolutionary Algorithm-Based OnLine Energy Management System for Plug-In Hybrid Electric Vehicles

Plug-in hybrid electric vehicles (PHEVs) have been regarded as one of several promising countermeasures to transportation-related energy use and air quality issues. Compared with conventional hybrid electric vehicles, developing an energy management system (EMS) for PHEVs is more challenging due to their more complex powertrain. In this paper, we propose a generic framework of online EMS for PHEVs that is based on an evolutionary algorithm. It includes several control strategies for managing battery state-of-charge (SOC). Extensive simulation testing and evaluation using real-world traffic data indicates that the different SOC control strategies of the proposed online EMS all outperform the conventional control strategy. Out of all the SOC control strategies, the self-adaptive one is the most adaptive to real-time traffic conditions and the most robust to the uncertainties in recharging opportunity. A comparison to the existing models also employing short-term prediction shows that the proposed model can achieve the best fuel economy improvement but requiring less trip information.

A Novel Blended Real-Time Energy Management Strategy for Plug-in Hybrid Electric Vehicle Commute Trips

Plug-in hybrid electric vehicles (PHEVs) show great promise in reducing transportation-related fossil fuel consumption and greenhouse gas emissions. A critical research topic for PHEVs is designing an efficient energy management system (EMS), in particular, determining how the energy flows in a hybrid powertrain should be managed in response to a variety of system parameters. Most of the existing systems either rely on a priori knowledge of future driving conditions to achieve the optimal but not real-time solution (e.g. Dynamic Programming strategy), or only upon the current driving situation to achieve a real-time but not optimal solution (e.g. rule-based strategy). Towards this end, we propose a Q-Learning based blended real-time EMS for PHEVs to address the trade-off between real-time performance and optimality. The proposed EMS can optimize the fuel consumption while learning the systems characteristics in real time. Numerical analysis shows that the proposed EMS can achieve a near optimal solution with 11.93% fuel savings compared to a binary mode control strategy, but a 2.86% fuel consumption increase compared to an off-line Dynamic Programming strategy.

An on-line energy management strategy for plug-in hybrid electric vehicles using an Estimation Distribution Algorithm

Plug-in hybrid vehicles (PHEVs) have great potential in reducing energy consumption and pollutant emissions, due to the use of electric batteries as another energy source. One of the critical considerations in PHEV development is the design of its energy management strategy, which determines how energy flows in a hybrid powertrain should be managed in response to a variety of system parameters. In this paper, we propose a generic framework of real-time energy management for PHEVs, where an Estimation Distribution Algorithm (EDA) is used for on-line (i.e., real-time) optimization of the power-split strategy. Different methods for controlling the battery packs State of Charge (SOC) are proposed and sensitivity analyses are conducted to evaluate their performance. Study results validate the effectiveness of the proposed methods and show promise for further field implementation.

Evolutionary algorithm based on-line PHEV energy management system with self-adaptive SOC control

The energy management system (EMS) is crucial to a plug-in hybrid electric vehicle (PHEV) in reducing its fuel consumption and pollutant emissions. The EMS determines how energy flows in a hybrid powertrain should be managed in response to a variety of driving conditions. In the development of EMS, the battery state-of-charge (SOC) control strategy plays a critical role. This paper proposes a novel evolutionary algorithm (EA)-based EMS with self-adaptive SOC control strategy for PHEVs, which can achieve the optimal fuel efficiency without trip length (by time) information. Numerical studies show that this proposed system can save up to 13% fuel, compared to other on-line EMS with different SOC control strategies. Further analysis indicates that the proposed system is less sensitive to the errors in predicting propulsion power in real-time, which is favorable for on-line implementation.

Energy Impact of Connected Eco-driving on Electric Vehicles

Transportation-related energy consumption and air quality problems have continued to attract public attentions. A variety of emerging technologies have been proposed and/or developed to address these issues. In recent years, electric vehicles (EVs) are deemed to be very promising in reducing traffic related fuel consumption and pollutant emissions, due to the use of electric batteries as the only energy source. On the other hand, recent research shows that additional energy savings can be achieved with the aid of Eco-driving system in a connected vehicle environment (e.g., Eco-approach at signalized intersections). However, most of the existing eco-driving research is only focused on the internal combustion engine (ICE) vehicles thus far. There is still lack of convincing evidence (especially with real-world implementation) of how these connected eco-driving technologies impacts the energy efficiency of EVs. To fill this gap, this chapter provides a real-world example of quantifying the energy synergy of combining vehicle connectivity, vehicle automation and vehicle electrification, by designing, implementing and testing an eco-approach and departure (EAD) system for EVs with real-world driving data.

A First-Order Estimate of Automated Mobility District Fuel Consumption and GHG Emission Impacts

A first of its kind, this study develops a framework to quantify the fuel consumption and greenhouse gas emission impacts of an Automated Small Vehicle Transit system on a campus area. The results show that the automated mobility district system has the potential to reduce transportation system fuel consumption and greenhouse gas emissions, but the benefits are largely dependent on the operation and ridership of the personal rapid transit system. Our study calls for more research to understand the energy and environmental benefits of such a system.

Integrated-Connected Eco-Driving System for PHEVs With Co-Optimization of Vehicle Dynamics and Powertrain Operations

In the past several decades, various types of technologies have been developed to improve vehicle fuel efficiency and reduce tailpipe emissions across different dimensions. For example, powertrain-related technologies improve fuel efficiency by optimizing the powertrain operations in response to different driving conditions (e.g., energy management system for plugin hybrid electric vehicles); another technology dimension lies in intelligent transportation systems (ITS), which improve vehicle fuel efficiency by optimizing the vehicle dynamics or speed under different traffic conditions (e.g., eco-speed harmonization and eco-approach and departure). However, very little effort has been made to investigate the combined benefit of integrating both powertrain and ITS technology dimensions together. In this paper, an integrated and connected eco-driving assistance system with co-optimization of vehicle dynamics and powertrain operations for PHEVs is proposed. To fully evaluate the performance of the proposed system at different vehicle automation levels, real-world driving data for different eco-driving technological stages were collected: uninformed manual driving, eco-driving with an in-vehicle advisory display, and an eco-driving system with automatic longitudinal control. The numerical analysis shows that the co-optimization is able to achieve on average 24% fuel savings for typical urban driving conditions.

Deep reinforcement learning-based vehicle energy efficiency autonomous learning system

To mitigate air pollution problems and reduce greenhouse gas emissions (GHG), plug-in hybrid electric vehicles (PHEV) have been developed to achieve higher fuel efficiency. The Energy Management System (EMS) is a very important component of a PHEV in achieving better fuel economy and it is a very active research area. So far, most of the existing EMS strategies just simple follow predefined rules that are not adaptive to changing driving conditions; other strategies as starting to incorporate accurate prediction of future traffic conditions. In this study, a deep reinforcement learning based PHEV energy management system is designed to autonomously learn the optimal fuel use from its own historical driving record. It is a fully data-driven and learning-enabled model that does not rely on any prediction or predefined rules. The experiment results show that the proposed model is able to achieve 16.3% energy savings comparing to conventional binary control strategies.

Intelligent On-Line Energy Management System for Plug-in Hybrid Electric Vehicles based on Evolutionary Algorithm

Energy management system (EMS) is crucial to a plug-in hybrid electric vehicle (PHEV) in reducing its fuel consumption and pollutant emissions. The EMS determines how energy flows in a hybrid powertrain should be managed in response to a variety of driving conditions. In the development of an EMS, the battery state-of-charge (SOC) control strategy plays a critical role. This paper proposes a novel evolutionary algorithm (EA)-based EMS with a self-adaptive SOC control strategy for PHEVs, which can significantly improve the fuel efficiency without knowing the trip length (in time). Numerical studies show that this proposed system can save up to 13% fuel, compared to other on-line EMS with different SOC control strategies. Further analysis indicates that the proposed system is less sensitive to the errors in predicting propulsion power demand in real-time, which is favorable for on-line implementation. Original publication: X. Qi, G. Wu, K. Boriboonsomsin and M. J. Barth, Evolutionary algorithm based on-line PHEV energy management system with self-adaptive SOC control, Intelligent Vehicles Symposium (IV), 2015 IEEE, Seoul, 2015, pp. 425--430.