| Developing new energy vehicles is becoming increasingly urgent to realize energy conservation and emission reduction. Taking advantages of both electrical vehicles and conventional vehicles, hybrid electric vehicles (HEVs)are quite promising to dominate the new energy vehicles market in the short term. To reduce the energy consumption of HEV efficiently, smarter power systems and configurations are being deeply studied by scholars and automotive companies at home and abroad. The power split hybrid electric vehicle(PS-HEV), taking planetary gear sets as the power coupling device, is now being the research foucs in the hybrid electric vehicles because that the power coupling device with planetary gear sets can decouple the torque and speed of the engine from those of the wheels. Besides, it can act as an electric variable transmission (EVT) so that the traditional transmission can be eliminated from the powertrain.Energy management strategy is the key to improving the energy conversation efficiency of the HEV powertrain. Power coupling device with planetary gear sets enriches the operation modes of the PS-HEV. Only with real-time and high efficient energy management strategy can PS-HEV fully improve the system energy conversion efficiency with the advantages of multi-modes and multi-freedoms. The PS-HEV energy management system(EMS) not only includes the discrete dynamic behaviors such as enabling/disabling mode switchings under various conditions, etc., but also the continuous dynamic behaviors constrained by its own physical properties.Consequently, the PS-HEV EMS shows distinct hybrid dynamic characteristics.Aiming at a new type of PS-HEV, this thesis studies its operation dynamics and modeling methods. Based on the hybrid system theory, the EMS hybrid dynamical model that can accurately reflect the system hybrid natures is further built. Ulteriorly, according to the established hybrid dynamic model, the energy distribution of PS-HEV is tried to be improved by optimizing the EMS. This research is assumed to provide a new theoretical basis and technical support for the design of PS-HEV.Firstly, the structure of the PS-HEV is described. The torque and speed characteristics of the planetary gear sets are analyzed. On this basis, the dynamical model and static model of the power coupling device are built.According to the structure features of the hybrid power system, the discrete operation modes of PS-HEV are decided, and the operation principle of feasible operation modes are analyzed with lever analogy to reveal the power flow and the torque relations between different power sources. The state transition relations between different modes are described. Combining the theoretical analysis and bench test data, models of the key components in the powertrain are established. By further introducing the dynamic model of the power coupling device, the simulation model, depicting the flow of energy and control signals of HEV, is developed to provide a basis for the vehicle performance evaluation, as well as the design and verification of the energy management strategy.Secondly, analysis of the PS-HEV hybrid dynamic behaviors and hybrid dynamic modeling of the EMS are conducted. According to the control features and requirements of HEV, the discrete event behaviors, continuous dynamic behaviors and the interactions between the two kinds of behaviors are revealed.Based on the hybrid system theory, the hybrid dynamic system of the EMS,described by a triunity, is built. The triunity contains the discrete decision level,continuous controlled level and an interface. With the extraction of the continuous and descrete information from the system by the interface, the communation between the discrete decision level and continuous controlled level is made. Simulation results under the China transit bus city driving cycle(CTBCDC) demonstrate that the hybrid control model structure could accurately reveal the hybrid dynamic behaviors during the PS-HEV energy management process, and realizes the mode swithing and torque distribution of different components. To further optimize the energy distribution within the PS-HEV, the hybrid control system of the PS-HEV should be further optimized.Thirdly, the PS-HEV operation modes are synthesized. Aiming at the synthetical operation modes, the control law generator in the PS-HEV hybrid controller is optimized based on the model predictive control algorithm. To predict the required torque of the driver, the driver model based on Markove Chain is designed with the data of the CTBCDC and actual transient driving cycle. In combination with the control targets and constraints of the energy management strategy, the optimal energy distribution problem of the PS-HEV is described as the constrained optimal control problem in the finite time horizon.The control effects and computation burden of both nonlinear model predictive control (NMPC) and linear time variant model predictive control (LTV-MPC)are investigated to explore the capability of model predictive control for practical applications. The NMPC takes the system nonlinear model as the prediction model. The algorithm based on equential quadratic programming(SQP) and that based on dynamic programming (DP) are studied to solve the nonlinear programming problem of NMPC in the finite prediction horizon,respectively. The LTV-MPC uses the linear model as the prediction model, and the energy management optimization problem is thus converted to the standard quardratic programming (QP) problem that is easy to obtain the optimal control sequence. Simulation studies and comparative analysis are conducted under various driving conditions to evaluate the performance of the designed optimal controller. The results show that the optimal energy distribution strategy based on LTV-MPC can improve the equavilent fuel economy of the PS-HEV efficiently. The improvement can be as much as 16.17% under the CTBCDC.Besides, the solving time, which is only about 45ms, is much shorter than those based on the SQP and DP algorithm. The LTV-MPC strategy shows good real time performance.Finally, vehicle test and hardware in the loop (HIL) test of the PS-HEV is conducted. The working characteristics of the two electric machines in the hybrid power system are studied through bench test, which provides basis for the parametric modeling of the electric machines. By means of vehicle test, the impacts of the planetary power coupling device on the vehicle fuel economy and dynamic performance are studied. The feasibility and accuracy of the established HEV simulation model based on Matlab/Simulink platform is also verified. On this basis, the HIL simulation platform is developed to test and analyze the performance of the control system. Test results of the HIL simulation validate the positive effects of the designed control system in improving the vehicle fuel economy and maintaining charging sustainability.The strategy is also proved to have the ability to meet the real time requirements.The study provides a basis for the vehicle test and product-level development of the designed optimal controller. |
PDF Full Text Request