Abstract: To estimate online health state of Li-Ion batteries in electric vehicles accurately, a method was proposed based on the parameters of a pseudo-two-dimensional model. Firstly, disassembling congeneric objective batteries and measuring their structural parameters using scanning electron microscopy, the method was arranged to get some unknown parameters based on genetic algorithm for an electrochemical model. Then, a new stoichiometry ratio-based battery positive capacity calculation was established to estimate the health state of battery. Considering the influence of aging on the stoichiometry ratio in the positive electrode, the estimation accuracy of health state was further improved. Finally, a battery aging dataset was used to verify the validity of the method. The results show that the proposed estimation method can achieve an accurate online estimation of battery health state in short dynamic loading.