最小二乘影响系数法的优化改进  

Improved Least Square Influence Coefficient Methods

作　　者：章璟璇[1] 唐云冰[1] 罗贵火[1] 

机构地区：[1]南京航空航天大学能源与动力学院,南京210016

出　　处：《南京航空航天大学学报》2005年第1期110-113,共4页Journal of Nanjing University of Aeronautics & Astronautics

基　　金：国防基础科研基金 (K1 60 1 0 60 71 1 )资助项目

摘　　要：阐述了柔性转子高速动平衡的最小二乘影响系数法和加权最小二乘法的基本原理 ,在此基础上提出了基于遗传算法的最小二乘法 ,并对考虑平衡质量受限时的最小二乘法进行简化修改后得到了限制配重的最小二乘法。比较分析这几种方法的适用性 ,进行了算例分析。结果表明 ,在对平衡配重量及平衡后的残余振动值要求不同时 ,应该分别采用不同的改进最小二乘法。其中基于遗传算法的最小二乘法 ,综合了以上改进最小二乘法的适用性 ,使平衡配重量和残余振动值均达到了优化。Basic principles of the least square influence coeffic ie nt method and the weighted least square method used in flexible rotor dynamic ba lancing are described. Restrained adjust mass least square method and least squa re method based on genetic algorithm are presented. Applicability of these me- thods is compared. The differences of these methods are proved by an example . Result indicates that different improved least square methods are used by cons idering the need of adjust mass and residual vibration value. Among these method s, the least square method based on genetic algorithm can be optimized by balanc ing adjust mass and residual vibration value.

关 键 词：转子动平衡 影响系数法 加权迭代 遗传算法 

分 类 号：V231[航空宇航科学与技术—航空宇航推进理论与工程]