Abstract
Experiment plays the prime method in soldering reflow profile forecasting. While its high cost and low efficiency makes the company hard to develop. According to the nonlinear relationship between the multi input and output, reflow profile input parameters setting method based on BP neural network and genetic algorithm(BPNN-GA) is proposed in the paper. Establish the finite element model of the PCB products with ANSYS software, and simulate the reflow profile. Temperature field in the PCBAs heat process is analyzed to help adjust to provide guidance for parameter settings partially. The result shows that the process of parameters set methods with BPNN-GA and simulating analysis with ANSYS is effective in establish the virtual test system. The obtained characteristic value meets product quality requirements well.