基于CVT的插电式混合动力及多目标优化,Plug-in HEV with CVT configuration, control, and its concurrent multi-objective optimization by evolutionary algorithm
International Journal of Automotive Technology, Vol. 15, No. 1, pp. 103 115 (2014)DOI 10.1007/s12239 014 0012 z
Copyright©2014KSAE/075 12pISSN 1229 9138/eISSN 1976 3832
PLUG-IN HEV WITH CVT: CONFIGURATION, CONTROL, AND
ITS CONCURRENT MULTI-OBJECTIVE OPTIMIZATION
BY EVOLUTIONARY ALGORITHM
Q. WANG* and A. A. FRANK
Department of Mechanical and Aeronautical Engineering, University of California Davis, CA 95616, USA
(Received 18 March 2011; Revised 8 November 2012; Accepted 11 June 2013)
ABSTRACT Plug-in Hybrid Electric Vehicle (Plug-in HEV) has dramatic improvements in fuel economy and emissionreduction. It is most important to decide its optimal configuration, energy management strategy, powertrain sizes, and controllogic parameters. For multi-objective optimization, we present a concurrent optimization methodology based on an optimalPlug-in HEV powertrain configuration with continuous variable transmission (CVT). The novelty is using evolutionaryalgorithm in conjunction with an instantaneous optimal energy management strategy. Simulation results indicate the proposedmethod can significantly reduce fuel consumption and emissions by simultaneously optimizing the propulsion systemparameters as well as the energy control parameters.
KEY WORDS:Charge depletion (CD), Charge sustaining (CS), Concurrent optimization, Continuous variable transmission(CVT), Fuel efficiency, Evolutionary algorithm, Plug-in hybrid electric vehicle (PHEV)
1. INTRODUCTION
Plug-in Hybrid Electric Vehicle (Plug-in HEV) (Frank,2005) has dramatic improvements in fuel economy andperformance. It is usually achieved by increasing fueleconomy and substitution of liquid fuel by a renewable andzero emission energy, electricity.
Most HEV optimization studies can be categorized intotwo types. First is the HEV strategy optimization thatfocused on developing optimal controls of the hybridpower sources (Baumann et al., 2000; Delprat et al., 2004;Francisco, 2002; Harmon, 2005; Saeks et al., 2002). Theother is the optimization of powertrain component sizing(Baumann et al., 2000; Delprat et al., 2004; Francisco, 2002;Harmon, 2005; Saeks et al., 2002), a key factor for achievingminimum fuel consumption, minimum emissions, bestperformance, as well as the lowest cost. This paperproposes a global approach that optimizes both powertraincomponent sizes and energy control logic parameters in aconcurrent manner, although it increases the complexityand computation effort. It is efficient in approaching theglobal optimum in a broader search space.
It is complex to decide the optimal propulsion system forconventional HEVs. The process becomes even moredifficult for a Plug-in Parallel HEV configuration withCVT (Continuous Variable Transmission) due to its twopower sources, CVT dynamics, and their effect on fuel*Corresponding author. e-mail: qwang@ucdavis.edu
103
economy, emissions and performance and the sensitivity togiven driving cycles.The main question is how to decide theoptimal configuration, energy management and powertrainsizes. This is a multi-variable problem related to hundreds ofparameters on system dynamics, powertrain models, andadvanced control strategies, in which the values of most ofthe variables depend on each other and constraint variables.This work presents an extensive study on the optimiza-tion problem for Plug-in HEV with CVT including itsoptimal configuration, control strategy, cost functionsanalysis, design constraints, and advanced probabilisticoptimization algorithms that lead to an accurate and optimalsolution for Plug-in HEV propulsion system adaptivelybased on real driving cycles.
2. RELATED WORK
Most HEV research is focusing on optimization usingadvanced control algorithms which lead to minimization ofthe energy circulation loss (Delprat et al., 2004; Francisco,2002; Harmon, 2005; Schurhoff, 2002). In addition to therule-based strategies (Lin et al., 2003; Meyr et al., 2000),there are optimal control (Delprat et al., 2004), fuzzy logiccontrol (Langari and Won, 2003), neural networks (Harmon,2005), and adaptive control (Saeks et al., 2002; Saeks andCox, 1999). The other category is HEV component sizingoptimization, and our literature survey classifies the recentapproaches into the following.
Heuristic parametric study, based on empirical data, hasresulted in many reliable implementations such as many
湖南大学 硕士学位论文 基于CVT的插电式混合动力电动...采用模糊控制方法对最优曲线控制策略进行了优化,并...1 Plug-in HEV结构示意图 2.1.2动力系统结构分析...
插电式混合动力汽车论文:插电式混合动力汽车 能量管理策略 粒子群优化 神经网络...【英文摘要】Compared with Hybrid Electric Vehicles (HEV), Plug-in Hybrid ...
插电式混合动力汽车论文:插电式混合动力汽车 能量管理策略 粒子群优化 神经网络...【英文摘要】Compared with Hybrid Electric Vehicles (HEV), Plug-in Hybrid ...
热门文档