Research on multi-objective optimization algorithm for structural stiffness of optoelectronic system and mirror quality of optical elements

View at publisher
Authors: Fan Zhou, Mingchao Wang, Liang Lei, Gang Cheng
Cited by: 0

Abstract

The performance of the opto-mechanical system is influenced by both structural stiffness and mirror quality of optical elements. Structural stiffness is the key index for the opto-mechanical system to resist external interference and maintain structural stability, and the mirror quality of optical elements is directly related to the imaging effect of the optical system. However, there is a mutual restriction between them, and improving the structural stiffness may increase the weight and cost of the system, thus affecting the mirror quality. In order to solve this problem, this paper proposes an optimization method of structural stiffness and mirror quality of optical elements based on multi-objective optimization algorithm. By constructing a multi-objective optimization mathematical model, the optimization objectives, design variables and constraints are defined, and the improved NSGA-II algorithm is used to optimize the design. Taking the primary mirror support system of a space telescope as an example, the optimization results show that compared with the initial design, the maximum deformation of the optimized system is reduced from 52 nm to 18 nm, the mirror roughness is improved from 0.82 nm to 0.55 nm, the first-order modal frequency is increased from 76 Hz to 103 Hz, and the total mass of the system is reduced from 12.3 kg to 11.8 kg. The optimization design not only significantly improves the stability and optical performance of the system, but also achieves the goal of lightweight, which provides an effective multi-objective optimization method for the design of optical and mechanical systems.

Fields of Study & Subjects

Adaptive optics and wavefront sensing Optical Systems and Laser Technology Advanced optical system design

Publication details

Journal / source
IET conference proceedings.
Volume
2025
Issue
34
Pages
266–271
Publication date
1 January 2026
Language
EN