A Joint Online Estimation Algorithm Of State Of Charge,State Of Health And State Of Power For Lithium Batteries

Lithium batteries have been widely used in transportation,power grids,mobile devices and etc.Nevertheless,in the automotive industry,due to the great diversity in operation conditions,lithium batteries,as one of the main power sources,not only require a timely response in these complex external conditions,but also a guarantee on the driver or the passengers' safety in arduous operational environment or hazards.During the process,an accurate estimation of battery states is urgently needed to ensure the safe operation of the battery and to supply a better battery management strategy.Common battery states consist of state of charge(SoC),state of health(SoH)and state of power(SoP).The three states respectively describe the battery's continuous discharge or charge capacity,residual operational life and instant power supply or recollection ability.So far,numerous estimation algorithms have been proposed to estimate these three state variables online but the limited computational capacity of the embedded system in automotive con-strains their applications.To tackle the conflict,the thesis proposes a joint algorithm,which estimates the three variables online basing on an optimized Randle battery model.As for the battery model,a normalized version of recursive least square algorithm is imple-mented on the identification of battery model's parameters,which successfully avoids the data overflow phenomenon caused by the recursive calculation of covariance matrix.The identified configurations are used to estimate the SoH of the battery directly.A novel extended Kalman filter algorithm(EKF)equipped with adaptive noise covariance matrix is implemented in the SoC online estimation.During the estimation,the open circuit voltage(OCV)is introduced as a state variable and is included in the recursive calculation process.To be followed,a multi-constraints peak power estimation algorithm based on the battery model and the state space in optimized EKF is proposed in the thesis.The calculation process has taken the voltage and current limits and the SoC limit into account.The calculation results together with the nominal power of the battery are applied in the SoP estimation.The accuracy of the three parts of algorithm is confirmed by simulative experiment.Eventually,experiments based on a battery monitor bench and BTS-5V300A verify the performance of the identification algorithm of the battery model and the SoC,SoH and SoP estimation results.The comparison between the results of simulative experiment and that of hardware based ex-periments verifies the accuracy of the estimation algorithm.