A method of combining Gaussian Process (GP) Surrogate model and Gaussian genetic algorithm is discussed to optimize the injection molding process. GP surrogate model is constructed to map the complex non-linear relationship between process conditions and quality indexes of the injection molding parts. While the surrogate model is established, a Gaussian genetic algorithm (GGA) combined with Gaussian mutation and hybrid genetic algorithm is employed to evaluate the model to search the global optimal solutions. The example presented shows that the GGA is more effective for the process optimization of injection molding.

Ontology stiffness of welding robot has an important effect on its positioning accuracy. Based on the parameterized model of the arm built with the software of ANSYS, this paper optimized the robot arm mass of the welding robot in method of first-order using the ANSYS parametric design language (APDL). The lightest mass of the arm was obtained on the premise of the stiffness requirement of the welding robot. From the results, we could assume that the first-order method was a scientific and accurate optimization method, which provided the theoretical direction of optimization as for the welding robot structure.