Hybrid Vehicle Transmission System Modeling And Control Optimization

ABSTRACT:Considering the great impact of fuel price increase and global warming trend, a growing demand on energy conservation and reduction of carbon dioxide emissions have been put forward world widely. The automotive industry has also been pressing to produce automobiles with better fuel economy and lower exhaust emissions. Compared with conventional vehicles, hybrid electric vehicles (HEV) are cleaner and more efficient. Based on the comprehensive studies of the domestic and foreign achievements, hybrid transmission systems are analyzed in this thesis.In the first place, a dynamic HEV model is derived. Considering the wide existence of nonlinear components in the transmission system, a linear-in-the-parameters model identification algorithm is derived, based on the net contribution criteria, so that both the structure and parameters can be obtained simultaneously and recursively. Simulation resultscomfirm its efficacy.According to the lever equivalent analog approach, a unified bond graph based analysis method is proposed and used for both the kinematics and dynamics modeling of a two-mode hybrid transmission system. Considering the unknown parameters of most components, a data based parameter identification algorithm is proposed, which usethe genetic algorithm (GA) and bond graph analysis results.Based on the analysisoffuzzy control and output feedback control algorithm, a fuzzy state feedback controller is presented and applied to the stabilization of a time-delay nonlinear system and tracking control. Compared with the conventional state feedback controller, it is easier for the proposed algorithm to select the controller parameters and with better robustness.Transformation of power withindifferent types, which is called "power circulation", happens frequently and influences the transmission efficiency greatly. To address this issue, a transmission speed ratio based control strategy is researched. Considering the efficiency of both engine and transmission system, the engine operation points for instantaneous optimal fuel economy is derived. The efficiency of the transmission system and vehicle fuel economy are significantly improved.To improve the mode switch and gear shifting impact of a hybrid vehicle equipped with an AMT, an optimal engine and clutch reference speeds generating algorithm is proposed. Gear shifting duration weight is used to balance the performance of gear shifting duration and smoothness. Numerical simulation results indicate that the balanced shifting performance can be achieved through using decoupled tracking control of the optimal reference speeds during the clutch engagement process.The nonlinear characteristic of the clutch and its driven systems determin that it is hard to get satisfied results with conventional control approaches. Therefore, a nonlinear optimal control algorithm is researched. By solving the state dependent Riccati equation (SDRE) each step, an optimal control law is derived. Hence, a well clutch displacement during the work condition switch and gear shifting process is guaranteed.Optimal cost function index matrices Q and R are usually obtained by empirical methods. Based on the concept of output covariance constraint, an optimal cost function index matrix calculation method is researched so that the stability and robustness of the optimal controller can be guaranteed simultaneously.A method that iteratively solving the SDRE online based on matrix vectorization and Newton gradient is derived. Considering the continuous of the system states, iterative results of each step are taken as the initial condition for the next step of SDRE solving. Compared with the conventional method, the computation load is dramaticlly reduced.In order to accelerate the clutch engagement process and also reduce the transmission output torque ripple, a coupled dual sliding mode control strategy is proposed. The two portion of the controller coordinate with each other and the torque change during the transition process are compensated by the motor. The controller can not only guarantee the fast mode switch, but also reduce the output torque vibration simultaneously.To sum up, through the research conducted in this thesis, further understanding of the key technologies of hybrid transmission system analysis can be achieved. The proposed analysis methods have strong adaptability and can provide reference for similar researches.