Energy Management Strategy For A Plug-In Hybrid Electric Bus Based On Model Predictive Control

The plug-in hybrid electric vehicles(PHEVs) gain focus recent years because of the abilities of energy saving and emission reduction. In this dissertation, a single-axis series-parallel PHEB powertrain is taken as the research object and the research mainly focus on the model predictive control(MPC) based energy management strategy for the PHEB.The potential operating mode and the power flow of the hybrid electric system are analysed. Considering the objective performance of the PHEB and the characteristics of the target driving cycle, the powertrain parameters of the series-parallel PHEB are matched and determined.Focus on the power flow and the fuel economy, the backward modeling method is selected. The PHEB backward simulation model is established, constructed by the the quasi-static models of the components. The rule-based strategy is formulated and the control performance is analyzed. The dynamic programming(DP) problem is formulated and solved, thus the global optimization energy management strategy is realized. The rule-based and DP algorithm based energy management strategy laid the foundation of the target performance for MPC based energy management strategy.The model predictive control(MPC) based PHEB energy management strategy is the key point of the research. By introducing the DP method to the MPC frame, The MPC problem is formulated and solved, and the method to reduce the computation complexity is researched. Three prediction methods are taken to predict the velocities and accelerations. The prediction effects are analysed and compared, and the multi-step markov prediction method is chosen for the further research. The MPC strategy is simulated. In order to prove that the receding horizon optimization strategy is effective, the real velocities within the prediction horizon are used to solve the optimization problem. Three methods are proposed to restrain the SOC, and the method based on the selection of control variables is selected for further research. The MPC simulation results are analyzed, and the operating area of the control varibles is optimized based on the simulation results. Through the comparison of MPC result with the results of the DP strategy and the rule-based strategy, it is certified that the control effect of MPC based strategy is much better than the ruled-based strategy and close to the global optimal control under DP based strategy. At last, the SOC trajectory based on the travel distance is established and ultilyzed to solve the MPC problem. Respectively based on the travel time and travel distance to establish the SOC reference trajectory, the simulation results indicate that the the fuel consumption per hundred kilometers under MPC based strategy both are reduced by 8.7% than the ruled based strategy result, and it’s close to the DP based strategy result.