The Research On The Method Of Power Lithium Ion Battery Pack State-of-health Estimation

The energy crisis and environmental pollution sounds a warning to people, so how to replace the traditional fuel vehicles become has been the subject of debate. Currently the Electric vehicles which apply power lithium-ion battery pack as a power source are highly concerned. In addition to the power lithium-ion battery, the battery management system is also underway. And accurately estimating the battery state of health (SOH) is the weak link in the present study, and directly impact to the Practicality of the lithium-ion a battery pack.The battery pack state of health (SOH) is a description of a slow and irreversible change in the process, with the increasing of the cycle frequency, the health status of the battery pack will be down trend. This decline is actually by the influence of the battery cell whose health status is seriously declining. Therefore by the battery management system to accurately estimate the health status of the battery pack and replace the single battery whose health status reaches a critical value, we can effectively extend the service life of the battery pack and cost the savings, has high economic efficiency. At the same time we also provide a reference for the powered battery state of charge (State of Charge, SOC) estimation and the battery balancing system. This is important for promoting the development of electric vehicles cause.Power battery SOC and SOH is two important state of the battery. However compared with the SOC, the research on SOH is lagging behind, and the estimation method is also not very mature. The SOH estimation method is divided into two types of traditional and modern. The traditional method is mainly based on the cycle of the battery charge and discharge test, by analyzing the characteristics of the battery to estimate battery SOH. The disadvantage is wasteful, long cycle test and not online real time measurement. This is not meeting the requirements of the electric vehicle. People study modern methods basing on the traditional methods. This method is based on a different battery models, application of fuzzy logic algorithm, neural network algorithm or Kalman filter algorithm to estimate the battery SOH, and received a good practical effect. And in these methods, the Kalman filter algorithm estimating battery SOH attracted much attention.The work is firstly analyzing the feature amount of the power battery SOH, the internal resistance of the battery and the rated capacity. The internal resistance of the battery is the analysis target. Then through analyzing the different battery cell model to identify the model which can sufficiently reflect the mathematical relationship of the battery internal resistance, and based on this to establish the Vmm model of the battery pack. Then do the test of the battery charge and discharge and obtain information about data, by least squares identification method to determine the model parameters, and to consider applying the method to the battery SOH estimates. On the one hand we can periodically detect the change of the SOH of the battery. On the other hand it can be used to calibrate the parameters of the battery model, and to improve the estimation accuracy. Finally, the battery model is built in Matlab and estimate the battery SOH with the application of the extended Kalman filter algorithm. First, try to use extended Kalman filter algorithm to estimate separately the amount of the internal resistance. Based on this, we can add the SOC factors, using double extended Kalman filter algorithm to estimate the internal resistance and the SOC of the battery.Apply the method of least squares and Kalman filter algorithm at the same time to estimate the battery SOH, and combine the static estimation and dynamic estimation to improve system reliability and the estimated accuracy.