Research On State Of Health Estimation Technology Of LiFePO₄ Battery With Strong Adaptability

Faced with the increasingly serious problems of resource shortage and environmental pollution,the state advocates vigorously developing new energy power generation and electric vehicles.LiFePO₄ batteries play an important role in these fields,with strong safety,high energy density and long life.As the core part of the stable operation of the power battery pack,the battery management system?BMS?also restricts the rapid development of the new energy industry.Among them,state of health?SOH?estimation is one of the important contents of the battery management system.Accurate SOH estimation can effectively exert the performance of the battery and make full use of the battery pack.In this paper,a 18650 LiFePO₄ battery was used as the research object to carry out the full life cycle accelerated aging experiment,fully considering the rate factor,and adding multi-rate charging characteristics test based on the conventional stress accelerated aging.According to the accelerated aging experiment process,until the end of battery life,the original external characteristics of the whole life cycle under variable stress conditions are obtained,which provides data support for subsequent battery SOH estimation.Based on the raw data obtained by the accelerated life aging experiment of the whole life cycle,data mining was carried out,and the capacity attenuation difference characteristics and multi-rate charging characteristics were quantitatively analyzed.The causes of the two characteristics were analyzed in detail from the microscopic point of the electrochemical mechanism.The general rule of aging difference characteristics and multi-rate charging characteristics of lithium iron phosphate battery was summarized.This paper focuses on the SOH estimation method with strong adaptability.Under different charging magnifications,the voltage range near the peak of the probability density function curve is regarded as the characteristic interval respectively.The dynamic sliding window scanning optimization method is used to determine the characteristic voltage interval,and the frequency of each voltage appearing in the interval is accumulated as the characteristic parameter P_f.The polynomial fitting method obtains the relationship between the characteristic parameters P_f and SOH,so that the aging characteristic table at different magnifications can be obtained,and then the SOH can be located by applying the constant current incomplete charging characteristics of the battery to be tested,and the estimation scheme can be adapted.The actual charging station or charging device has different charging rates,and can also adapt to the battery aging difference,and achieve a high-precision SOH estimation with a dynamic sliding window scanning optimization strategy.In order to verify the accuracy and effectiveness of the SOH estimation scheme proposed in this paper,the Freescale MC9S12XEG128 is used as the control core to build the BMS experimental platform.Linear Corporation's LTC6804 and AMS's AS8510 are used for battery voltage and current sampling respectively,and based on this hardware.The platform was written to complete the relevant experimental platform test,and the validity and accuracy of the proposed SOH estimation method were verified from multi-angle experiments.