腐蚀条件下基于信息更新的飞机结构剩余寿命预测

发布时间：2019-04-17 21:47

【摘要】：论文针对民航业所面临的在役飞机结构腐蚀问题,探索在役飞机结构的腐蚀发展规律,研究在役飞机结构剩余寿命的预测方法和最优腐蚀维修决策,对确保飞机结构的安全运行、延长使用寿命和降低维修成本具有一定的实用价值。论文首先基于在役飞机典型结构的实际腐蚀检测数据,分析飞机结构腐蚀频数随检出时间与机龄的动态发展规律。建立在役飞机典型结构腐蚀频数和主要气象环境因素的主成分回归模型,讨论主要气象环境指标与飞机结构腐蚀频数的内在相关性。其次,建立在役飞机典型结构的灰色马氏链剩余寿命预测模型,并进行残差修正提高模型预测精度。以隔框结构实际腐蚀检修数据进行算例分析,预测结果验证了模型的可信性。再次,给出基于贝叶斯公式的飞机典型结构腐蚀深度的合理分布函数的选取方法,建立基于等维递补和贝叶斯更新思路的飞机结构剩余寿命动态预测模型。算例结果表明隔框结构腐蚀深度的合理分布为Logistic分布,而且更新的预测结果更符合腐蚀维修工程实践。最后,建立以飞机结构损伤打磨容限为机会约束的,单位维修成本期望值最小为目标的混合整数规划(MIP)模型。设计改进的分支定界法,求解最优预防性检测次数N*和检修时间间隔T*。论文研究成果可为民用飞机结构腐蚀检修和维护提供技术支持。
[Abstract]:Aiming at the corrosion problem of aircraft structure in service, this paper explores the corrosion development rule of aircraft structure in service, and studies the prediction method of remaining life of aircraft structure in service and the optimal corrosion maintenance decision. It is of practical value to ensure the safe operation of aircraft structure, prolong the service life and reduce the maintenance cost. Firstly, based on the actual corrosion detection data of typical aircraft structures in service, the dynamic development rule of aircraft structure corrosion frequency with detection time and aircraft age is analyzed. The principal component regression model of typical structure corrosion frequency and main meteorological environment factors of aircraft in service is established, and the intrinsic correlation between the main meteorological environment index and the corrosion frequency of aircraft structure is discussed. Secondly, the grey Markov chain residual life prediction model for typical structures of aircraft in service is established, and the residual error correction is carried out to improve the prediction accuracy of the model. The reliability of the model is verified by the numerical example analysis based on the actual corrosion maintenance data of the spacer structure. Thirdly, the selection method of reasonable distribution function for the corrosion depth of typical aircraft structures based on Bayesian formula is given, and the dynamic prediction model of aircraft structure residual life based on the idea of equal-dimensional complement and Bayesian updating is established. The numerical results show that the reasonable distribution of the corrosion depth of the spacer structure is Logistic distribution, and the updated prediction results are more consistent with the practice of corrosion maintenance engineering. Finally, a mixed integer programming (MIP) model is established, which takes the damage tolerance of aircraft structure as an opportunity constraint and the minimum expected value of unit maintenance cost as the goal. An improved branch and bound method is designed to solve the optimal number of preventive detection N * and the interval of maintenance time. The research results can provide technical support for civil aircraft structural corrosion repair and maintenance.
【学位授予单位】：中国民航大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：V267

【相似文献】