Optimization Of Lithium Batteries Quick Charging For Electric Vehicle

The 13 th Five-Year Plan Draft(2016-2020) clearly puts forward that our next task is to promote the development of low carbon transportation. Simultaneously, the missions also include putting the promotive plan of new energy vehicles into practice and improving the level of industrialization of electric vehicles. The final task is to set up the complete science and technology system and a better industrial chain of electric vehicle power system in 2020. As the core component of electric vehicles, power battery has been restricting the rapid development of electric vehicles for a long time. It is a heated topic that how to improve the safety, efficacy and convenience of power battery.Firstly, referring to the existing battery test specifications at home and abroad, this article compares the basic properties of a variety of polymer lithium ion batteries and takes the price and technology maturity into account. Furthermore, the article selects the LiFePO4 power battery as study object and lays a good groundwork for battery modeling.Secondly, by analyzing and comparing several present and widespread battery models which include the mechanism of electrochemical model, neural network model, equivalent circuit model and so on. The article tries to establish the optimization model for power battery based on PGNV equivalent circuit and uses Hybrid Pulse Power Characteristic test(HPPC) to obtain model parameter. In addition, the article applies to the least square method to identify the model parameter and analyzes the change regulation about the open circuit voltage, ohm internal resistance and time constant in the state of charge(SOC). According to relevant condition in《USABC Battery Test Procedures Manual》, The article also assesses the simulation accuracy of PNGV battery model and verifies that the model is suitable for LiFePO4 power battery.Finally, from the analysis of the PNGV simplified equivalent circuit mode,it is established that the mathematical model expression which is based on time domain. In the face of limitation of the steepest descent algorithm such as the small size of iteration step and slow convergence speed, the article proposes the steepest descent optimization algorithm based on the golden section law. Then the piecewise constant current charging has been optimized through using this algorithm. The parameters for the piecewise constant current charging termination also has been confirmed. Meanwhile, it reduces current falling gradient and scope of the piecewise constant current charging. After an optimization and adjustment, the actual charging current curve of power battery is more close to the acceptable charging current curve, which shortens the charging time and improves charging efficiency.