摘要
轨道板运输车设计多依赖于经验,机身结构有较大的刚度、强度余量。为了保证结构的可靠性,充分发挥材料的承载性能,提出一种基于ISIGHT的轨道板运输车可靠性优化方法。该方法在原始设计时考虑不确定性因素对结构可靠性的影响,并将其集成到多学科优化软件ISIGHT中,构建神经网络近似模型,通过蒙特卡洛模拟技术,对结构进行6sigma质量分析;在质量分析的基础上引进优化模块,最后采用多岛遗传算法对目标函数寻优。结果表明,原始设计的轨道板运输车结构不满足可靠性设计要求;优化后,在满足结构可靠性的条件下,轨道板运输车结构质量减少了1 900.58kg,与优化前质量相比自重降低了13.89%,优化效果显著,对轨道板运输车的结构设计及改进有重要的实际意义。
The design of rail plate carrying vehicle mostly depends on experience,and the fuselage structure has large stiffness and strength allowance.In order to ensure the reliability of the structure and give full play to the bearing capacity of the material,a reliability optimization method of rail plate carrying vehicle based on ISIGHT was proposed.In the original design,this method considered the influence of uncertain factors on the reliability of the structure,and integrated it into the multidisciplinary optimization software ISIGHT to build an approximate model of neural network.6 sigma quality analysis was conducted on the structure through monte carlo simulation technology.On the basis of quality analysis,the optimization module is introduced,and finally the objective function is optimized by the multi-island genetic algorithm.The results show that the structure of the original rail plate carrying vehicle does not meet the reliability design requirements.After optimization,under the condition of meeting the structural reliability,the structural mass of the rail plate carrying vehicle was reduced by 1 900.58 kg,and the dead weight was reduced by 13.89%compared with that before optimization.The optimization effect was significant,which had important practical significance for the structural design and improvement of the rail plate carrying vehicle.
作者
王道成
WANG Dao-cheng(China Railway No.4 Engineering Group Co.Ltd.,Hefei Anhui 230041,China)
出处
《装备制造技术》
2019年第2期69-73,共5页
Equipment Manufacturing Technology
关键词
轨道板运输车
可靠性优化
神经网络近似模型
rail plate carrying vehicle
reliability optimization
neural network approximation model