Abstract： Technological parameters in fused deposition modeling three-dimensional printing are coupled and the forming process is a non-linear forming process. A large number of modeling parameters affect the quality of the product precision in fused deposition modeling of three-dimensional printing. In order to clarify the effects of various parameters on the forming precision and improve the precision of three-dimensional printing, the wavelet neural network prediction model of product precision was build by using the Matlab software. The arithmetic was designed and the samples were acquired by fused deposition modeling experiment. The training samples were used to train the network to accomplish the mapping relation between the input and output of the network. The test samples were used to verify the performance of the trained network. Simulation results indicate that the prediction model has sufficient accuracy. The prediction model of product precision is feasible and valid in theory and in practice. The wavelet neural network method was used to model the relation between the processing parameters and the product precision in fused deposition modeling of threedimensional printing. The difficult problem to create the accurate mathematics model was solved.