Abstract: Based on the gasification reaction model of coal chars, the hybrid neural network model for simulation of coal char gasification process was set up, connecting with BP neural network parameter estimator. The results show that the hybrid neural network model is quite feasible to describe the gasification process of coal chars and the two parameters (A, the ratio of the carbon with activity and the total carbon in coal char; Rr, the unit mass reaction rate of the carbon with activity in coal char), which can not be measured directly in experiments.