In this paper, straw and rice bran is the object of the molding process. It analyzes the impact of straw and rice bran mixture ratio and temperature on the main indicators of forming fuel. Based on the LS-SVM, multi-objective optimization function and the model of the straw and rice bran mixed-compression molding process have been established. Also the model and molding process has been tested and optimization with the experimental data. The results show that the model prediction and experimental values of the average maximum relative error is 1.59%, with good simulation results. Meanwhile, when the proportion of rice straw is 55.9% and the molding temperature is about 129.07°C, the optimal value is very close to the experimental maximum value and the maximum relative error is 2.86%.