摘要
为了讨论不同优化算法的效果差异,利用有限元分析技术,针对典型1-0-1支承方案的双盘转子系统进行建模仿真,在分析其动力学特性的基础上,联合多学科优化软件ISIGHT,将两转盘位置作为优化变量,以1阶临界转速在10%内的变化为约束条件,采用不同的优化算法:进化算法(EVOL)、多岛遗传算法(MIGA)、邻域培植遗传算法(NCGA)、第二代非支配排序遗传算法(NSGA-Ⅱ)以及Pointer方法,最小化过1阶临界转速时两转盘的最大振幅,得到两转盘的最优位置.在高速柔性模拟转子试验器上进行了瞬态试验,结果表明,NSGA-Ⅱ只需要计算240个点,就可以使两转盘振幅分别下降76.94%和67.42%.因此,NSGA-Ⅱ是最适合该类转子系统的优化方法.
In order to analyze the performance difference of several optimization algo- rithms, a model of two-disk rotor system with typical 1-0-1 support scheme was established and the dynamic characteristics were analysed by finite element analysis technology. On this basis, the two-disk rotor system model was integrated to the multidisciplinary design optimi- zation software ISIGHT. The positions of the two disks were selected as the optimization variables, the first order critical speed with the variation range of 10% was set as the con- straint, and the optimization objective was to minimize the amplitudes of the two disks cross- ing the first order critical speed with several different algorithms including evolutionary opti- mization （EVOL）, multi-island genetic algorithm （MIGA）, neighborhood cultivation genetic algorithm （NCGA）, non-dominated sorting genetic algorithm Ⅱ （NSGA-Ⅱ） and Pointer. The transient experiment was carried out on a high-speed flexible rotor mockup. Experimen- tal results indicated that NSGA-Ⅱmade the two disks＇ amplitudes decreased 76. 940/4 and 67.42% respectively with the calculation of 240 points. It can be concluded that NSGA-Ⅱ was the best optimization algorithm for the rotor system of this type.
出处
《航空动力学报》
EI
CAS
CSCD
北大核心
2016年第1期65-71,共7页
Journal of Aerospace Power
基金
航空科学基金(20112153019)
关键词
双盘转子
优化算法
多目标优化
第二代非支配排序遗传算法
ISIGHT软件
two-disk rotor
optimization algorithm
multi-objective optimization
non-dominated sorting genetic algorithm Ⅱ
ISIGHT software
