摘要
针对民机研制阶段直接维修成本(Direct Maintenance Cost,DMC)优化问题进行研究,建立可靠性、维修性协同优化(Collaborative Optimization,CO)模型。模型以DMC最小化为目标,以可靠性参数MTBUR、MTBF和维修性参数MTPM为设计变量。为了求解模型,将遗传算法与反向传播神经网络组合,通过BP神经网络预测输出,再采用遗传算法寻求最优解。根据设计变量确定网络结构为3-5-1,采用实数编码,通过选择、交叉、变异求解最优解及对应设计变量值。最后以冲洗包和压缩机为例通过Matlab仿真研究,验证协同优化在民机研制中的可行性。
In order to study the direct maintenance cost(DMC)optimization of civil aircraft during the development stage,the Collaborative Optimization Model in regard to reliability and maintainability is established.Minimizing the maintenance cost is set as the target,and reliability parameters MTBUR,MTBF and maintainability parameter MTPM are chosen as the design variables.An improved Genetic Algorithm(GA)combined with Back Propagation(BP)Neural Network is designed to get the optimal solution.Firstly the BP Neural Network is used to get the prediction,and then the GA is used to acquire the optimal value.The network structure is defined as 3-5-1 according to the design vector.Individuals are coded by real and the optimal solution with corresponding design vector values are got through selection,crossover and mutation.Finally,flushing set and compressor are taken as examples,and the effectiveness and applicability of this model are verified through the simulation.
作者
贾宝惠
吴婧
JIA Bao-hui;WU Jing(College of Aeronautical Engineering,Civil Aviation University of China,Tianjin 300300,China)
出处
《机械设计与制造》
北大核心
2019年第6期264-267,共4页
Machinery Design & Manufacture
基金
民用飞机直接维修成本分析与控制技术研究(MJZ-2014-Y-61)
面向维修设计的分析与适航验证平台(MHRD20160105)
关键词
直接维修成本
可靠性
维修性
协同优化
遗传算法
研制阶段
Direct Maintenance Cost
Reliability
Maintainability
Collaborative Optimization
Genetic Algorithm
The Development Stage
