Research On Modeling And SOC Estimation For LIFEPO4 Battery Of Electric Vehicle

Under the National Natural Science Foundation of China（51265039,51261024）and the Jiangxi Provincial Science and Technology Program funded projects（No.20141BBE50021） and battery manufacturers（Original Energy Technology）, according to the dynamic characteristics of lithium iron phosphate in different working conditions, the paper establish RC correction model that has a specific relationship of the battery’s internal state and external characteristics, study in-depth the extended Kalman filter algorithm based on RC modified model, use algorithm to estimate the state of charge for lithium iron phosphate, this algorithm based on RC modified model has better applicability, provided help for the design of the battery management system, the research mainly include the following contents:1. Discusses background and significance of the research. The status of domestic research about the battery model and state of charge to estimate are summarized. The structure, working principle and advantages of lithium iron phosphate battery was introduced in detail, manufacturing process of lithium iron phosphate battery are given. Based on this, puts forward the main contents and the main innovation points of this paper.2. On the analysis of the lithium iron phosphate battery performance under different working conditions, including the basic characteristic of the battery, electronic, safety performance, environmental applicability and safety performance, on the basis of a large number of test experiment was carried out, including the voltage characteristics under the standard charge and discharge experiments, the ratio of charging and discharging voltage characteristic experiment, capacity ratio characteristics, the correlation between the open circuit voltage and capacity test experiments under different charge and discharge rate, the temperature of the battery capacity characteristic experiment and the cycle life characteristics of battery. The test result shows that the lithium iron phosphate battery at the same temperature, the greater charge ratio, the floating phenomenon is more obvious; The lower the discharge current, the greater the battery output power; When the environment temperature is higher, the main factor affecting the output capacity of the battery is not discharge ratio, but by the environmental temperature. The environment temperature is lower, the main factor affecting the output capacity of the battery is discharge rate, and low output capacity of the battery in low temperature environment; Lithium iron phosphate batteries has lower self-discharge and have long cycle life. These valuable conclusions for subsequent model battery and provides an important support.3. Discusses the three typical battery model of lithium iron phosphate batteries, namely the electrochemical model, the thermal model and performance model, especially for the equivalent circuit model is analyzed, and points out the deficiency existing in them. On this basis, aiming at the problems existing in the traditional battery RC model and its deficiency in practical applications, such as traditional RC model of the battery internal resistance as a fixed value, and only consider the battery, the relation between the current and open circuit voltage of battery SOC and ignoring the self-discharge, cycle times, and the battery polarization effect and other factors. This paper proposes a new RC correction model, this paper puts forward models have more advantages than the traditional RC model, correction of RC model not only consider the influence factor of traditional RC model, also give full consideration to the battery temperature, internal resistance, current, self-discharge, cycle times and battery SOC polarization effect to the model. Finally, the RC correction model put forward by the application of the least squares method for parameter identification, the identification results can be for subsequent battery charged state to estimation important support to provide.4. Discusses the definition of battery charged state, on this basis, further discussed the internal resistance and temperature and charging and discharging rate on the influence of the battery SOC estimation and the strength of the impact factor. In view of the classic battery charged state estimation algorithm of using RC correction model is established, based on extended kalman filter is proposed battery charged state estimation method, and the initial value of recursive equation are discussed and analyzed. Compared with the traditional method, the proposed method has good dynamic performance, is not sensitive to initial value error, and has good inhibition of measurement noise and other features.5. In order to verify the accuracy of the model and the effectiveness of the battery charged state estimation, here, has set up a test platform for the battery, including hardware experiment platform and the battery test software. RC correction model of the battery are simulated by Simunlink. Electricity, the experimental results show that the constant exile, HPPC circulation and UDDS different working conditions, the maximum error of the measured and the average error is very small, the maximum error is not more than 0.1385 V, the average error value is less than 0.0197 V, relative maximum error less than 4.2%, put forward correction model has a high estimation precision.At the same time, in the experimental test platform, based on the RC charged state estimation method of correction model of EKF and experimental verification. The experimental results show that the exile in constant power, constant current pulse discharge, CYC_UDDS and CYC₁051_prius four different working conditions, the average error is less than 2.87%, the maximum error is less than 4.89%, therefore, put forward the charged state estimation method has high accuracy and good anti-jamming capability.