Abstract: To overcome the difficulty that an accurate mathematic model is hard to obtain,a batch process model based on fuzzy neural network with nonlinear fuzzy rule consequence is proposed.Then a parameters learning algorithm based on Lyapunov method with global convergence is also presented with rigorous proof.The proposed algorithm possesses the high approximation and better self-learning ability,thus it provides a new way for the modeling of batch processes.Lastly,to verify the efficiency of the proposed algorithm,it is applied to a benchmark batch process.The simulation results show the efficiency and potential practical value of the proposed model.