Method and System for Screening the Best Carrier Material in Lithium-Sulfur Battery Based on Machine Learning

Lithium-sulfur battery is the most common type of rechargeable battery in portable electronic devices. These batteries have high energy density and can discharge at a very high rate, which makes them very suitable for portable electronic devices. However, these batteries have several safety problems, which may cause catastrophic failure during discharge or overcharging. The most common problem of lithium-sulfur batteries is sulfation, which occurs when the anode (lithium side) is saturated with sulfuric acid and prevents further charge transfer from the cathode (sulfur side). This paper provides a method and system for selecting the best carrier material in lithium-sulfur battery based on machine learning, including: optimizing the structure of AB2 two-dimensional layered carrier material; Set up a variety of polysulfides, calculate the adsorption energy using density functional theory, and build a data set corresponding to the adsorption structure and adsorption energy; The structure in the data set is calculated by atomic local chemical environment to obtain the structure characteristics of the data set; The depth neural network is trained based on the structural characteristics of the data set, and the adsorption energy of any configuration and any site is predicted according to the trained depth neural network; The migration learning algorithm is used to adjust the parameters of the depth neural network for the preset AB2 type two-dimensional layered carrier material to adsorb polysulfide; The depth neural network for the preset AB2 type two-dimensional layered carrier material to adsorb polysulfides is trained, and the adsorption energy of any configuration and site is predicted according to the trained depth neural network, and the adsorption capacity is measured according to the average value.