Study On Key Techniques For Plug-in Hybrid Electric City Bus

Plug-in hybrid electric vehicles (PHEV) can continuously run20km to80km solelydriven by the power battery. Beyond that range they can be driven by engine and the electricpower resource together. And besides, PHEV can use the220V50Hz power resource torecharge its batteries directly. On the other hand, the typical city bus driving cycle in China isthat start-stop and acceleration-deceleration work conditions are occurred frequently, drivingvelocity is usually low, driving range is often short, and the engines have to idle for a longtime with lower fuel-burning efficiency and heavy emission pollution. Therefore, consideringthe battery performance bottlenecks and due to the advantage that PHEV takes into accountthe advantages of hybrid electric vehicles (HEV) as well as of purely electric-driven vehicles,PHEV is suitable for the city bus work conditions; it is one of the optimal solutions for thetransition from traditional fuel-consumed city bus to the pure electric city bus. There are manyof unsolved key technical problems for PHEV so far which result that few real vehicles run inthe domestic and international city bus market, such as transmission selection and matching,energy management and control for the whole vehicle, brake energy recovery, and vehicledesign and manufacturing etc.The research works are conducted for algorithm and application innovations focusing onthe key techniques on transmission selection and matching design, vehicle controllingstrategies, battery management systems and regenerative braking energy feedback.1) As for transmission selection and matching issue, to fulfill the requirements of thetypical city bus driving cycle in China, based on the nondominated sorting genetic algorithm-Ⅱ(NSGA-Ⅱ), a design scheme is proposed for the transmission structure determination, forthe parameter matching and optimization, and for the component selection. Firstly thetransmission system is chosen a parallel structure design; the transmission system parametersare matched. And then, based on an improved NSGA-II multi-objective genetic algorithmeach gear transmission ratios are optimized. The concreted models and parameters of thecomponents for the transmission are proposed finally. The results show that the scheme canmeet the dynamic requirement, besides that it can significantly improve the economicindicators of the vehicle at the same time. 2) As for vehicle energy management and control, another kind of rule-based controlstrategy is proposed, it is proved that it can ensure vehicle dynamic and improve fuel-burningefficiency. Firstly, one CAN bus based distributed control system is built on. Secondly,control logics on work condition transition and control strategies on each work condition aremake out. The work conditions consist of pure motor start, pure engine start, running withpure electric, running with pure engine, charging, running co-driven by motor and engine, andregenerative braking. And then, simulation models are formulated such as the models forengine external characteristics, models for motor/generator torque, power and efficiencycharacteristics, models for battery charge and discharge, and models for state of chargeprediction. The system model for the whole powertrain is also drawn out. At finally, thedynamic performance and fuel economical performance are evaluated based on the fueleconomy simulation program we developed.3) As for battery management system, a V-flow based rapid development method for thestate of charge (SOC) estimation module in battery management system is proposed. Firstly,based on Matlab/Simulink software platform a battery model and a SOC estimation algorithmbased extended Kalman filter are formulated, and their function are verified via offlinesimulations. And then, the Simulink blocks of the battery model and the estimation algorithmare translated into targeted codes by Matlab real-time workshop (RTW) automatic codegeneration tools. And based on pre-development MPC555embedded platform rapidprototyping developments are completed. Finally the codes verified by the rapid prototypingare downloaded into the prototype vehicle and are calibrated online there, the SOC estimationfunction module validation is finished at this actual test. The results demonstrate that theextended Kalman filter based SOC estimation algorithm improves the prediction accuracy,and the V-flow can result in improvement on development efficiency.4) As for regenerative braking energy feedback, a hybrid braking control strategy is putforward which can maximize the regenerative brake energy. Simulation models for thefriction braking sub-system and the regenerative braking sub-system are established on.Considering the restrictions of the ECER13braking regulations, and of the characteristics ofthe motor, the lithium-ion batteries and the transmission, simulation experiments areconducted for the proposed braking force distribution and control strategy under all typical braking operation conditions. The results demonstrate that for all the typical braking operationconditions the composite braking system can effectively guarantee the braking safety andmaximize the recovery braking energy at the most degree.5) According to China’s hybrid city bus licensing authorization specification, actual roadtesting experiences are make on the developed plug-in hybrid city bus prototype vehicle,test indices include fuel consumption amount per100km, pure electric-driven mileage,maximum speed, acceleration time, braking distance, etc. The results demonstrate that all theindices are qualified, and the fuel economy advantage is especially obvious. It indicates thatthe results of this research have practical value.