Abstract: The state of charge(SOC) and state of health(SOH) of lithium-ion battery are important parameters to be estimated during the operation and maintenance of battery energy storage systems. In order to reliably estimate battery states, the simple recurrent unit(SRU) in deep learning is used to achieve joint estimation of SOC and SOH. Firstly, a SOC estimation model based on SRU is established by taking advantage of SRU in dealing with timing problems. Then, the input form of data unit is introduced to the model, and the sample data containing battery aging information is used to train the model, so that the trained model can achieve SOC estimation at any degree of battery aging. Finally, the SOH of the battery is estimated by mining the aging information contained in the SOC estimate output by the model. The experimental results show that the proposed method can accurately estimate SOC and SOH of batteries, and has good applicability to different types of batteries.