6σ robust and optimal design of micro-motion platform based on kriging model

Abstract

The two performance indexes of output displacement and inherent frequency are closely related to the design variables of the micro-motion platform. To address the problem that the design variables have certain fluctuations and the high nonlinearity between the design variables and the performance of the micro-motion platform, a multi-objective robust optimization method with high computational efficiency and high fitting accuracy is proposed. Based on the kriging agent model, the high-precision approximate functions of the key design variables with the output displacement and the inherent frequency are established by the Latin hypercube experimental design, respectively. With the maximum stress as the constraint, Monte Carlo simulation, 6σ robust design concept, and nondominated sorting genetic algorithm II (NSGA-II) multi-objective genetic algorithm are integrated to optimize the structure of the micro-motorized platform. The results show that compared with the traditional deterministic optimization method, the 6σ robust design considering parameter uncertainty effectively reduces the fluctuation of the performance of the micromotion platform and improves the resistance of the platform performance to disturbance under uncertainty.