基于多目标遗传算法的雷载计算机隔振系统优化设计  

Optimization Design of Vibration Isolation System for Torpedo Borne Computer Based on Multi-objective Genetic Algorithm

作　　者：张溢文[1,2] 尹韶平[1] 王志杰[1] 郭君[1] 吴培明[1] 崔鑫山[1,2] 

机构地区：[1]中国船舶重工集团公司第705研究所,陕西西安710077 [2]水下信息与控制重点实验室,陕西西安710077

出　　处：《水下无人系统学报》2017年第2期57-63,共7页Journal of Unmanned Undersea Systems

摘　　要：为解决复杂力学环境下雷载计算机的振动隔离问题,提出了一种基于UG和ANSYS Workbench的联合优化设计方法。在UG中建立了隔振系统的参数化几何模型,并导入ANSYS Workbench建立有限元模型。计算了原始设计方案的模态参数,通过与模态试验结果对比,验证了计算模型的正确性。随后依据隔振设计理论确定了优化目标函数,使用多目标遗传算法对隔振系统进行了优化设计。通过对比可知,优化方案较原始方案的最高阶固有频率降低了12.9%,频率间隔降低了79.6%,系统的有效隔振频率增加了39.1 Hz,达到了良好的设计效果。To solve the vibration isolation problem of torpedo borne computer in complex mechanical environment, this paper proposes a joint optimization design method based on UG and ANSYS Workbench. A parametric geometry model is established by UG and imported into ANSYS Workbench for finite element model ing. Modal parameters of the origi-nal design scheme are calculated and compared with the modal test data to prove the correctness of the calculation model. Then the optimal object ive function is determined based on vibration isolation theory, and optimization design of the vibration isolation system is carried out using mul ti-objective genetic algorithm. Compared with the original design, the optimization scheme reduces the maximum natural frequency by 12.9% and the frequency interval by 79.6%, and increases the effective vibration isolation frequency by 39.1 Hz.

关 键 词：雷载计算机 优化设计 多目标遗传算法 隔振 模态理论 

分 类 号：TJ630.3[兵器科学与技术—武器系统与运用工程] TB115.2[理学—数学]