Reliability-based design optimization of a cradle-type double rotary table for 5-axis machining center

In practical engineering, the inevitable presence of certain uncertainties in target systems due to differences in material properties, inaccurate characterization of load conditions, and manufacturing errors. Therefore, this paper considers the influence of random factors and defines the ability of the rotary table to resist deformation under load as the reliability of the rotary table system. We propose a reliability-based design optimization (RBDO) model that comprehensively considers mass, mechanical performance, and carbon emissions. First, a mechanical performance model and a carbon emission model for the cradle-type double rotary table have been established. Considering the assembly relationships within the internal structure of the rotary table, the sensitivity indicators for each component’s structural parameters are obtained through sensitivity analysis, and these indicators are used to guide the optimization process. During the optimization process, an adaptive Kriging surrogate model is employed to reconstruct the functional relationship between the structural parameters of the rotary table and its performance. Given the high dimensionality and wide range of the design variables, a multi-stage optimization method has been proposed to overcome the impact of training data density on the fitting accuracy of the surrogate model and improve the convergence effect of the optimization. The optimization results indicate that the multi-stage optimization method can achieve better outcomes in a shorter time compared to traditional optimization approaches. Moreover, compared to deterministic design optimization, RBDO significantly enhances the reliability of the system, providing a new solution for the optimized design of the double rotary table and other complex mechanical structures.