Lithium-ion battery state of health estimation (SOH) is important for safe battery use, and traditional SOH estimation methods are not accurate enough to meet the needs. Due to this, proposed an improved particle swarm optimization algorithm and long and short term memory network lithium battery SOH estimation (DNPSO-LSTM) method A novel particle swarm optimisation (NPSO) method is improved and the number of neurons and learning rate parameters of the LSTM are optimised at and; Finally, a NASA standard lithium battery dataset was used for training and testing. The experimental results show that the SOH estimation metrics RMSE, MSE, MAE and MAPE of the method outperform the traditional method.