基于重要性测度降维的液压管路稳健优化设计

To research the robust optimization problem of aircraft local hydraulic piping system in uncertain environment, a dimension-reduction pretreatment method based on importance measure is presented. The interval model is used to describe the uncertainty of the random variable distribution (i. e., design variables), and the importance measure based on variance is introduced to measure the contribution of design variables to the optimization goal. The design variables which have large effect on the optimization goal are then screened out, and the optimization model is simplified. The proposed method can reduce the dimension of optimization problem and thus the cost of calculation dramatically. A numerical example is provided to illustrate the rationality and feasibility of the proposed method, and the robust optimization problem of the local aircraft hydraulic piping system is solved with the method.