Study Of Energy Management Strategy Of Range-Extended Electric Vehicles

Range-Extended Electric Vehicle(REEV)combines the features of pure Electric Vehicles(EV)and traditional Hybrid Electric Vehicles(HEV).REEVs typically offer good All Electric Range(AER)and thus good short range fuel economy,while at the same time supporting uncompromised total driving range with its Auxiliary Power Unit(APU).Because of these merits,REEVs have been widely studied by research facilities and auto makers around the globe.Energy Management Strategy(EMS)is one of the main focuses among these studies.It affects to a great extent the energy-saving potential of REEVs.However,there are still a number of issues existing in the current study of EMS for REEVs,namely: Current studies mostly use fix driver models and specific drive cycles,and often neglects the complex and constantly changing real-world traffic conditions and driver driving styles.As a result,they tend to exhibit lackluster adaptiveness in real-world applications.In order to take real-world traffic conditions and driver driving styles into consideration to better formulate the EMS,and to improve the real-world application potentials of this reasearch,the study of real-time EMS for REEVs with consideration of driver driving characteristic information,traffic condition information and corresponding optimization theories is very much necessary.This paper studied the EMS for REEV to improve its fuel economy,adopting multiple methods to take driver characteristics and traffic conditions into consideration,and achieved the real-time application of the EMS on a real vehicle with good results.Specific works of this paper are as follows:(1)Parameter matching,optimization,modeling and simulation of the powertrainCompleted parameter matching of the powertrain of the REEV according to design requirements,optimized powertrain parameters with genetic algorithm,optimized engine operating points with efficiency optimization,built forward models of the REEV and run simulations,validated the model and corresponding parameters,thus setting the basis for further study of the EMS.(2)Studied the EMS for REEV based on Stochastic Dynamic Programming(SDP)First,transformed the actual driving power requirement into a discrete Markov process and found the transition matrix of actual power requirement under given driving cycles,thus enabling the EMS to take driver drving styles and traffic conditions into account when calculating optimal energy distributions.Then,introduced traffic cost coefficient that reflects the difference between actual energy cost and simulation energy cost,calculated the values of the coefficient based on standard drive cycles,and applied it to the instantaneous cost function at every instant,thus enabling the EMS to take traffic conditions into account when calculating energy cost.Last,as a global optimization algorithm,simulation results of the SDP EMS serve as reference for consequent studies of the real-time EMS.(3)Proposed the real-time EMS based on Equivalent Consumption MinimizationStrategy(ECMS)with consideration of driving styles and traffic conditionsFirst,proposed to use questionares and multi-variable statistics to guide the design of membership functions of the driving style fuzzy-logic identifier.Achieved a comparatively accurate identifier,and control logic is chosen based on identification results.Then,proposed to acquire and predict traffic information on different levels according to the richness of usable traffic information,and assigned or calculated the value of equivalent factor of ECMS in real-time.To be specific,when floating car data is rich,future vehicle speed is predicted using the nearest neighbor method based on floating car data,and the equivalent factor is calculated in real-time according to future energy cost prediction;when floating car data is lacking,traffic information is identified using a trained neural network,and the equivalent factor is assigned in real-time according to the identification result.Last,the identified driver driving style,extracted traffic information,predicted future vehicle speed and ECMS are employed together to complete the EMS.Corresponding simulations were conducted.(4)Hardware-In-the-Loop(HIL)simulations and vehicle testsBuilt HIL simulation platform and proved the real-time abilities of the proposed EMS and controller.Performed driver driving style identification test to verify the accuracy of the dirving style identifier;performed vehicle road tests to verify the accuracy of the traffic condition identifier and predictor;and performed dynamometer tests to prove the energy distribution and energy saving abilities of the proposed EMS in real vehicle applications.