基于Kringing和贝叶斯优化的飞机翼梁结构可靠性优化  

Reliability optimization of aircraft wing beam structure based on Kringing and Bayesian optimization

作　　者：张佳琪 苏长青 王昱 ZHANG Jiaqi;SU Changqing;WANG Yu(College of Safety Engineering,Shenyang Aerospace University,Shenyang 110136,China;National Defense Key Subject Laboratory of Digitization of Aeronautical Manufacturing Process,Shenyang Aerospace University,Shenyang 110136,China)

机构地区：[1]沈阳航空航天大学安全工程学院,沈阳110136 [2]沈阳航空航天大学航空制造工艺数字化国防重点学科实验室,沈阳110136

出　　处：《沈阳航空航天大学学报》2025年第1期66-72,共7页Journal of Shenyang Aerospace University

基　　金：国家自然科学基金(项目编号:51005228);辽宁省教育厅项目(项目编号:JYTMS20230228);国家市场监督管理总局重点实验室开放基金(项目编号:2023KF003)。

摘　　要：为了提高飞机的飞行性能及飞行安全性,根据适航规定确立飞机翼梁结构的实际工况。通过设计翼梁结构参数、构建综合目标函数、设置约束条件、构建可靠性优化数学模型,完成对翼梁结构的可靠性优化设计。采用Kriging与贝叶斯优化的可靠性双层优化求解策略,对翼梁结构进行求解。结果表明,随着目标函数调用次数的递增,翼梁结构的质量呈现出明显的下降趋势,不同工况下翼梁结构的位移也随之减少,最终得出翼梁结构的质量最小值和不同工况下的最小位移。研究结果为飞机翼梁结构优化设计提供了理论支撑。The practical working conditions of the wing beam structure were established according to the airworthiness regulations to improve the flight performance and safety of the aircraft.The reliability optimization design of the wing beam structure was completed by designing the wing beam structure parameters,constructing the comprehensive objective function,setting the constraint conditions and constructing the reliability optimization mathematical model.The reliability double-layer optimization strategy of Kriging and Bayesian optimization was adopted to solve the problem of the wing beam structure.The results show that with increasing the calls of the objective function,the mass of the wing beam structure shows a significantly decreasing trend,and the displacement of the wing beam structure under different working conditions is also reduced.Finally,the minimum mass of the wing beam structure and the minimum displacement of the wing beam structure under different working conditions are obtained.These research results provide theoretical support for the optimization design of aircraft wing beam structure.

关 键 词：翼梁结构 Kriging代理模型 贝叶斯优化 质量最小值 最小位移 可靠性优化 

分 类 号：V222[航空宇航科学与技术—飞行器设计]