| Extended-Range Electric Vehicle (E-REV) is a kind of Plug-in Hybrid Electric Vehicle(P-HEV), which is an interim one between a traditional Hybrid Electric Ve hicle and ElectricVehicle. With the installation of Range Extender, the E-REV could achieve longer mileagethan the Electric Vehicle (EV). Hence, it has been regarded as the most desirable interimproduction of EV. Three issues are studied in this dissertation: vehicle energy managementoptimization, battery life degradation study and battery life optimization for E-REV. Themain contributions are as follows.Firstly, an E-REV optimal energy management law is extracted. The E-REV vehiclebattery energy distribution, with different driving conditions, trip distance, control strategiesand battery characteristics, is optimized off-line via the global dynamic programmingalgorithm. And then, the theoretic optimal discharging trajectory of the battery characterizedby the State of Charge(SOC) is obtained. The optimized result shows that when the E-REVenergy usage is at the lowest cost, the SOC trajectory can be approximately equivalent to aline segment, and the slope is the ratio of the battery discharging window with mileage.Based on the conclusion described above as well as the SOC trajectory of E-REV underdifferent control strategies, the―Capacity-distance ratio(CDR)‖is proposed to characterizethe E-REV vehicle control strategy mode and battery energy usage, which equals the ratio ofbattery SOC discharging window divides its trip distance correspondingly. CDR can be usedto research the affection of E-REV control strategy on its battery life.Secondly, an E-REV battery life prediction model is developed. The following threeparts are integrated:1) determination of the model structure: according to the internal agingmechanism of the battery, a model structure which considers the calendar life and cycle lifesimultaneously is determined; the battery life accelerating factors are chosen as temperature,battery SOC and its depth of discharge(DOD), and the acceleration models of the factors are determined; the battery life model structure is determined by integration of the abovemodels.2) extracting general parameters which are fitting parameters of acceleration models:to the E-REV battery in this research, the general parameters are extracted from thelong-term life data of the same type of battery.3) specific parameters identification whichcharacterize the individual property of the battery, the specific parameters are identified fromthe short-term life test data of the E-REV battery in this research. With the same modelstructure and general parameters, a fast parameter calibration approach of battery lifeprediction model is proposed for other similar battery life analysis.Thirdly, a life prediction model of E-REV battery under variable vehicle workingconditions is proposed. For the purpose of comparing battery degradation resulting fromvariable accelerating factors profiles, an equivalent substitution method is used to translatethe variable factors to a constant equivalent value to predict the battery life. Through theSOC trajectory of the battery, a function model between vehicle driving conditions andequivalent life accelerating factors is established, from which a fast estimation of the batterylife can be realized under the prerequisite of the E-REV vehicle working condition. In thevehicle pre-development stage, it can be used for estimating battery life and vehicleperformance when the battery achieve its end of life(EOL), which could also provide areference value for E-REV battery parameters oversizing margin when considering batterylife degradation.Fourthly, a manage approach of an E-REV battery life optimization is established. Itcontains vehicle parking environmental selection, the timing choice of battery charging afterparking, control strategy selection under low CDR, dynamic discharging windowmanagement and thermal management of the battery system. The first two aspects areinfluenced by user habits, and recommendations to optimize battery life are given to usersvia voice prompts after parking. The last three aspects are dependent on the OEM activeintervention management of the battery life, and it is realized by control strategydevelopment, control strategy parameter adjustment and battery pack design. Compared witha constant static battery discharging window, a dynamic modification method of the E-REV control strategy parameters is proposed based on the fade battery capacity, it can realize thedynamic discharging window management, which will optimize battery life, improve theefficiency of the battery and vehicle system, and increase the pure electric driving mileageduring the whole vehicle life time.Lastly, an E-REV vehicle control system and its hardware in-the-loop test platform aredeveloped, providing verification platform for energy management strategy design and themanagement of the battery life, which will also monitor the battery life state and verify itslife management results all the time.The main innovations are as follows:(1) The law of E-REV optimal energy management is extracted and the―Capacity-distance ratio(CDR)‖is proposed to characterize the E-REV energy managementstrategy mode, which can be used to research the affection law of E-REV control strategy onbattery life degradation.(2) A prediction model of E-REV battery life under variable vehicle working conditionsis developed，and E-REV battery life can be fast pre-estimated under the E-REV targetworking condition, which will help estimate battery life and vehicle performance when thebattery achieves its EOL state or provides a reference value for E-REV battery parametersoversizing margin when considering battery life degradation.(3) An active management approach of E-REV battery life optimization is proposedoriented to the whole usage time of the battery. In which, with―dynamic SOC concept‖, theE-REV battery discharging window is managed based on its fade capacity, which optimizesbattery life of the whole usage time and utilizes effectively of the battery capacity. |
PDF Full Text Request