Latest research progress of lithium batteries-Part 3
4 Prediction and development of battery state of charge
The prediction of the battery state of charge is the top priority of the power management system. The accurate detection of the battery state of charge can not only provide the user with the supply status of the battery, but also is the basis of battery charge and discharge management and energy balance control. The inaccurate prediction of the state of charge of the battery may lead to the occurrence of overcharge and overdischarge of the battery, thereby affecting the service life of the battery; the unsatisfactory prediction of the SOC of the battery will also cause the power management system to lose control of energy distribution. Cause a waste of energy. Therefore, accurate SOC prediction can extend battery life and reduce battery cost to a certain extent.
To improve the accuracy of real-time online estimation of SOC, it is necessary to conduct in-depth research and summary on measurement methods, methods, battery models and estimation algorithms. In addition, the traditional battery fuel gauge also requires the battery to be fully charged and fully discharged to update the battery capacity, but this rarely happens in real applications, which may lead to greater measurement errors. Therefore, it is difficult to accurately estimate the remaining battery capacity and working time during the battery operating cycle.
At present, the remaining battery power prediction methods include ampere-hour measurement method, open circuit voltage method, fuzzy control method, Kalman filter method, neural network method, impedance spectroscopy method, and linear model method. The simpler and practical method is the combination of ampere-hour measurement method and open circuit voltage method. The method that is being studied more is the combination of ampere-hour measurement method and Kalman filter algorithm. The rest of the methods are only in the theoretical research stage, and it will take some time before they are applied.
Another drawback of the Coulomb counting algorithm is that the total capacity of the battery can be updated only when the battery is fully discharged immediately after it is fully charged. However, portable device users rarely fully discharge the battery. Therefore, the actual power may be greatly affected before the update is completed. reduce. The disadvantages of this method are: (1) the initial battery value SOC0 cannot be provided; (2) the current measurement is inaccurate, which makes the estimation of SOC over time Accumulation produces cumulative error, which deviates from the actual value; (3) When estimating SOC, battery efficiency coefficient η (including temperature influence coefficient ηT and charge-discharge rate coefficient ηi),which requires a lot of tedious experiments to get the estimated value of η, but the generality of practical application cannot be guaranteed under experimental conditions.
In recent years, the extended Kalman filter has made substantial progress [7-9]. The extended Kalman filter is considered to be the best adaptive algorithm based on recursive estimation. When the Kalman filter algorithm is applied to battery SOC estimation, the estimation accuracy will change with the changes in the electrochemical characteristics, temperature and service life of the battery. The estimation error can be improved by repeated measurement of parameter values, but this repeated The iterative process is very time-consuming. Therefore, at present, only under controllable experimental conditions, the Kalman filter algorithm can be applied to the SOC estimation of single cells [10-11].
At present, there are many SOC chips based on the coulomb measurement method, such as TI's bq2026, MAXIM's DS2786, DS2781/2788 and so on. In addition, the latest release for single-cell Li+ battery packs is that the company uses the latest ModelGauge m3 algorithm to estimate SOC. This is also the only coulomb in the industry that avoids the mutation correction in the traditional coulomb measurement method. It can be used to measure the voltage of up to 12 batteries in series.
The prediction of the SOC of lithium batteries is an important part of battery balance management, but it is also a very difficult problem. The theoretical or applied chips mentioned above are only in the preliminary research stage and need to be further explored. Moreover, most of the chip products mentioned above are used in high-end portable electronic products such as notebook computers and digital cameras, and their applications in power batteries are relatively relatively low. less.
The above research facts indicate that the safety and plasticity problems of lithium batteries can be effectively solved by studying the polymer electrolyte of lithium batteries. Secondly, in-depth research and development of new materials that can increase the capacity and charging speed of lithium batteries can solve the obstacles of lithium battery capacity and long charging time. Finally, due to the particularity of the lithium battery itself, a balanced protection circuit must be designed for the battery pack to meet the long-life use of the battery, so that the voltage of the individual cells in the battery pack will not be discrete with the charge and discharge cycle, ensuring the battery Will not overcharge and discharge. It is believed that with the continuous in-depth research on lithium battery materials and protection circuits, the popularization and application of lithium batteries will soon be realized.
 Gui Changqing. Power Battery [M]. Beijing: Mechanical Industry Press, 2009: 301-302.
 Qu Weiping. Broad prospects and development obstacles of lithium batteries [J]. Power Supply Technology Application, 2009, 12 (8): 6.
 GODIL A. A charger capable of choosing fast or slow charging [M]. Electronic Design Technology, 1999, 25(1): 5.
 SMITH K M, SMEDLEY K M. Engieering design of lossless passive soft switching methods for PWM convers[J]. IEEE transaction on Power Electronics, 2001(16):336-344.
 TOBISHIMA S I, TAKEI K, SAKURAI Y, et al. Lithium-ion cell safety[J]. J Power Sources, 2000 (90): 188-195.
 GITZENDANNER R, PUGLIA F, MARTIN C R, et al. High power and high energy lithium-ion batteries for under-water application[J]. J Power Sources, 2004, 136(4): 416.
 PLETT G L. Extended Kalman filtering for battery management sys- tems of LiPB-based HEV packs part I-III[J]. J Power Sources, 2004 (134): 252-261.
 VASEBI V, PARTOVIBAKHSH M, BATHAEE M T, et al. A novel combined battery model for state-of-charge estimation in lead-acid batteries based on extended Kalman filter for hybrid electric vehicle applications[J]. J Power Sources, 2007 (174) ：30-40.
 HAN J, KIM D, SUNWOO M. State-of-charge estimation of lead-acid batteries using an adaptive extended Kalman filter[J]. J Power Sources, 2009 (188): 606-612.
 LEE S, KIM J J. State-of-charge and capacity estimation of lithium- ion battery using a new open-circuit voltage versus state-of-charge [J]. J Power Sources, 2008, 185(2): 1367-1373.
 LEE J, NAM O, CHO B H. Li-ion battery SOC estimation method based on the reduced order extended Kalman filtering[J]. J Power Sources, 2007 (174): 9-15.