纯电动汽车磷酸铁锂电池SOC在线估算系统设计与实现-控制工程专业论文.docx

Abs Abstract II II Abstract The SOC (battery of charge), from which we can not only obtain the remaining capacity of the battery but also estimate the remaining mileage, is an important parameter of the Battery Management System, hence it is of great significance to the development of the pure electric vehicle. There are following reasons for the SOC to be accurately estimated very difficultly. Firstly, the SOC can’t be measured directly. Secondly, the working environment of the battery is complex. Thirdly, it is characteristic of highly nonlinear. Lastly, the estimation of the SOC is also influenced by many factors, such as the voltage, the temperature, the charging and discharging current, the measurement accuracy of the sensor and so on. In this paper, as for the SOC estimation of the LiFePO4 battery, the main research content is as follows: Based on not only analyzing the research status and the shortcomings of the technology of the SOC estimation at home and broad but also comparing the advantages and disadvantages among different methods, we can conclude that no single estimation method can estimate the SOC accurately, instead, only by a combination of various methods can we get better results. According to the relationship between the SOC estimation and the model, we prove that the PNGV model can reflect the characteristics of the LiFePO4 battery accurately. Furthermore, we also establish the PNGV model of the LiFePO4 battery, meanwhile, we carry out the parameter identification of the PNGV model. The implementation of the SOC estimation algorithm: The SOC estimation system of the battery is highly nonlinear. By the comparison between EKF and UKF, we know that the UKF is superior to the EKF on the estimation of the SOC. This paper proposes a estimation method of combining the UKF and the open circuit voltage method based on the PNGV model with the advantages of many different kinds of algorithms. We use the ARBIN battery test cabinet to obtain the actual batt