Research On Integrated Optimization Method Of Hybrid Power System For Plug-in Hybrid Electric Vehicle

With the stern challenges of energy crisis and environmental pollution, plug-in hybrid electric vehicles(PHEVs) have attracted widely attention in automobile industry all over the world. The high specific energy battery pack used in pure electric vehicles and the high specific power of lithium-ion batteries used in hybrid vehicles are difficult to independently satisfy the power and energy needs of the dual-energy sources system. This thesis will focus on the hybrid power system consists of power batteries and ultracapacitors in PHEVs, aim to get a higher specific power and specific energy combined system, and to extend the cycle life of the hybrid power systems. The detailed work has been carried out.The determination of the topologies of the hybrid power system has two crucial issues. Firstly, the capacitor of the support assistant and the react capacity of the energy recovery, Secondly, to reducing the impact of high current and improve the cycle life of the battery and the whole system. To systematic analysis and evaluation the common used topologies of the hybrid power system and the control mode based on whether DC/DC is participate or not. Finally, to clarify the topology of the hybrid power system based on the power battery system and DC/DC inverter and then in parallel with the ultracapacitor. And also build the dynamic numerical dynamic simulation for the hybrid power system.Build the test platform and designed the test program for the components in hybrid power system, and established the experiment database for the power batteries and ultracapacitors. Build the dynamic simulation model for the two components, and the parameters optimization of models was proposed based on the genetic algorithm. Set up a composite plug-in hybrid electric bus models with Cruise software for the plug-in hybrid electric bus equipped with battery-ultracapacitor hybrid power system.First proposed the hybrid power system state estimation concept. Based on the decoupling of the energy and power characteristics for the hybrid power system, the fusion state of charge(SOC) estimation method based on the fuzzy optimization and extended Kalman filtering algorithm, and the state of charge estimation for the ultracapacitor based on the real-time forecast of operation voltage was proposed. Validation test was carried out for the uncertain operation conditions, aging state and inaccurate SOC initial value. The validation results shows that the maximum estimation error of the batteries was limited less than 2%, which would provide an accurate decision factors for the energy management strategies for the hybrid power system.Based on the building the optimization model of the hybrid power system, the integrated optimization method was proposed based on the particle swarm optimization(PSO) and dynamic programming(DP) algorithm. The application of the PSO and DP are aim to determine the parameters of the hybrid power system and the optimal control ratio with different parameter set. Set the energy losses of the hybrid power system, minus the charge and discharge current rate peak value and average value of the power battery as the optimization object, systematic evaluate and determine the optimal parameters set for the hybrid power system.Based on the analysis of operation mode for PHEV, energy distribution management strategy was derived and identified under different operating modes. The analysis of energy need and power need based on the typical cycle and dynamic index was carried out. And the energy management optimal strategies for the hybrid power system was proposed. To determine the parameters such as battery capacity, ultracapacitor capacity and voltage based on the Chinese Typical City Driving Cycle(CTCDC) and China- World Transient Vehicle Cycle(C-WTVC), DP method was used to determine the best working performance and optimal control rate between power batteries and ultracapacitors. To analysis the energy-saving mechanism of hybrid power system and extraction of optimal control law, to design the energy optimization management strategy. Simulation results between the traditional logic threshold strategy and improved control rules shows that integrated optimization method based on the extraction of energy management strategy can effectively reduce the energy consumption of 6.48% in hybrid power system. Meanwhile, comparison and analysis are simulated for the charging and discharging current of the single battery power system and hybrid power systems. Simulation results show that the current in the single cell system will changed with the load current of power demand, and exhibit the phenomenon of alternating positive and negative. The battery current range within the limited range in the hybrid power systems, reducing the peak current impact on power battery, also the impact of the battery current rate of charge and discharge rate for the battery cycle life was qualitative analysis and discussed.To elucidate the problem of the validation of energy management of the hybrid power system, build the hardware-in-the-loop simulation platform based on the xPC target real-time simulation platform. The PHEV Simulink® model was downloading to the target machine, and analysis the real-time needed power of the electric motor, and send it to the Soaring electronic load instrument and to control the real-time charging or discharging for the hybrid power system. To prove the state of charge estimation algorithm and energy management strategy for the hybrid power system. The bench test results show that the proposed SOC estimation method has higher estimation accuracy, and the energy management strategy is feasible.The state of charge estimation method of the power batteries proposed by this thesis based on the fusion algorithm between fuzzy logic control optimization and extended Kalman filtering algorithm, solve the problem of higher accuracy state of charge estimation, to reach the accuracy state estimation for the hybrid power system. Provide a strong theoretical support for the energy management in plug-in hybrid electric vehicles. The proposed integrated optimization algorithm combined the particle swarm optimization and dynamic programming method solve the problems of the parameters matching and the collaborative optimization, updating and calibration problems for the hybrid power system, and the research results have a strong theoretical research significance and engineering application value in multi-energy system used in plug-in hybrid electric vehicles.