【作者】 张军红；

【导师】 韩景龙；

【作者基本信息】 南京航空航天大学 ， 固体力学， 2010， 博士

【摘要】 飞机气动弹性及其优化设计是飞机研制中的重要内容之一。已有工作主要基于传统的确定性方法,而实际工程中,不确定性因素无处不在,且不确定程度参差不一,将直接导致其实际性能和设计值相差较远。而气动弹性优化方案更需要具有稳健性,并能在设计阶段就将各种系统参数不确定因素纳入设计范围。本文在全面回顾了传统气动弹性颤振优化分析方法的基础上,对包含不确定性的机翼气动弹性优化方法进行了系统性的研究和探索。分别研究了包含随机不确定因素的颤振优化设计和包含区间不确定因素的颤振优化设计。采用响应面法实现了基于随机不确定性的机翼结构颤振优化设计分析,和基于区间不确定性的机翼结构不确定性分析。数值算例证明了方法的可行性和高效性。本文主要贡献和创新包括如下几个方面:1、根据模糊控制不依赖于被控对象的精确模型,对参数变化不敏感,具有很强鲁棒性的特点,提出一种模糊逻辑控制器,对包含间隙的二元机翼极限环振荡进行控制,并对其响应机理进行探讨和研究。数值仿真结果表明,模糊控制器能够有效抑制非线性极限环振动,使系统迅速达到稳定,提高系统颤振速度。2、考虑气动弹性系统设计参数的随机不确定性,提出一种基于响应面法,以机翼结构的颤振性能为设计目标的可靠性优化方法,并用复合材料层合板翼模型验证其可行性和正确性。该方法能够高效率地对包含随机不确定性的机翼颤振问题进行优化。3、考虑气动弹性系统参数随机不确定性,提出一种基于神经网络代理模型的机翼结构颤振可靠性优化设计方法。该法效率高,节约计算成本,能够有效解决复杂机翼结构不确定性颤振优化问题。4、针对工程问题中不确定因素统计数据信息不完备、待设计目标不具备函数表达式问题,提出了基于Kriging代理模型、包含区间不确定性参数的机翼结构稳健性优化设计方法。该方法给出了稳健性优化模型,并将其转换为易于实现确定性形式。该方法不需要考虑不确定性参数的分布型式,也不要求目标函数或约束函数连续性、可导性等,对数学模型要求不高;更方便工程设计人员选择对目标稳健性和约束稳健性的设计偏好。最后对本文工作进行了总结,并展望了未来研究方向。更多还原

【Abstract】 One of the most important steps in airplane design process is aerodynamic analysis and optimization on wing structures. Now most of work which had been researched is based on traditional deterministic optimization methods. But in real engineering projects, uncertainty exists everywhere inevitably. This directly leads to the performance of products are far from design. The aeroelastic design schemes of aeroelastic systems should have robust stability, and all kinds of uncertainties should be taken into consideration from the beginning of optimization design process.Based on systematic review of the traditional optimization design methods for aeroelastic design, research and exploration on the optimization methods about aeroelastic design containing uncertain parameters is performed in this thesis. The research work on optimization methods for aeroelasic problems of wing structures which contain interval uncertainties and random uncertainties is performed separately. Numerical examples show the feasibility and efficiency of those methods. The main work of this thesis is listed as follows:1. Fuzzy controller has good robustness, which is not sensitive to uncertainty of disturbance, without the requirement of the exact mathematic model. Based on such characteristics, a novel controller based on fuzzy control strategy is designed to suppress the LCO vibration of typical wing section with control surface freeplay. Simulation results indicate that this method is valid and the fuzzy controller has good performance.2. Considering various random uncertainties exist in aeroelastic system, a novel reliability-based method, using the response surface method, is developed to perform optimization of wing structures with the object of maximum flutter speed. An example of plane composite material wing structure is presented. The results show the validity and correctness of this method. This method can complete the optimization of wing flutter problem containing uncertain parameters efficiently.3. A novel robust optimization method is developed which is based on neural network surrogate model for flutter optimization problem including random uncertain parameters. The method can solve the wing structure flutter optimization problem with high efficiency and economizes computational cost.4. For engineering problems which lack of statistics information of uncertainty and without explicit formulation of design objects, a novel robust design optimization method using interval analysis is present based on Kriging surrogate model of wing structures. The robust optimization model is built based on the method of non probability robust optimization theory, and is changed to the deterministic form which is suitable for using the traditional optimization methods. This method has the advantages that it has less requirements on uncertainty and mathematical. This method is very convenient for engineering analysist to show their preference whether on objective performance or constains performances.At last, the research work of this dissertation is summarized and the prospective of further research is discussed.更多还原