Hybrid intelligent algorithm and CAD secondary development technology in parameter optimization design of transmission gear set

The traditional production of automobile transmission gears mostly adopts the empirical design method, resulting in problems such as oversized transmission volume, increased cost, and reduced driving comfort. Based on this research, the transmission gear set parameter optimization system was built through CAD drawing platform. After analyzing the gear parameter design principles, the genetic algorithm was improved through nonlinear programming, and the back propagation neural network was integrated to calculate the gear strength. In the simulation experiments, the method built in this study reduced the average volume of the original transmission from 71513 to 55064 mm³, and its volume of the transmission decreased by 23%. At the same time, the maximum displacement amplitude of the transmission under the gear optimization of the improved algorithm constructed in this study was between 0.9 and 1.1 um, and the vibration amplitude was smaller than other algorithms. In addition, the deformation degree of the back-propagation neural network optimized by genetic algorithm on the x, y and z axes was 0.7, 0.25, and 0.95 um, respectively, in the deformation analysis of the transmission connecting bearing, which was smaller than other optimization algorithms. The genetic algorithm-back propagation (GA-BP) algorithm was constructed to reduce maximum noise to 81.1 dB and achieved high driving comfort. Meanwhile, the average deformation of the three-axis was 0.505, 0.092, and 1.158 um, which was lower than other algorithm models. The experiment shows that the improved hybrid intelligent algorithm constructed in this study has good performance in the optimization of transmission gear parameters, reducing volume and deformation noise of transmission and noise.