Distillation is a unit operation process that uses the different volatility of various components in the mixture to separate or purify. The key problem has always been how to improve the separation effect of rectification. However, there are complex coordination, constraints and conflicts between the objective function and the input parameters of the rectification system, and it is difficult to obtain the optimal operating conditions by applying traditional control methods. With its powerful computing power and adaptive learning ability, the artificial neural network can establish the nonlinear correspondence between the objective function and multiple independent variables without relying on the mathematical model, which provides a powerful tool for the optimal operation and design of the distillation column. This article will introduce the basic principles, typical network models and development history of artificial neural networks, and summarize the application and research progress of artificial neural networks in distillation.