应用遗传算法的非线性转子-轴承系统参数辨识

Parameter Identification of Rotor-Bearing Systems with Non-linear Behavior by Using Genetic Algorithm

为了识别非线性油膜力作用下的转子-轴承系统的不平衡量,给出了一种适合参数识别的非线性油膜力表达式,并引入遗传算法来识别非线性油膜力的动力系数和不平衡的大小和位置。首先运用达朗伯原理结合里茨法建立了非线性油膜力作用下的多自由度转子-轴承系统的运动方程,进而利用遗传算法对非线性油膜力参数和不平衡量进行识别。结果表明:该方法能够有效识别转子的不平衡量,并能达到较高的识别精度,可以用于无试重动平衡。

For the purpose of identifying the magnitude of unbalance of rotor-bearing systems, subjected to non-linear oil film forces, an expression, representing non-linear oil forces, and suitable for identifying parameters, is set forth. A genetic algorithm is further introduced for identifying the dynamic coefficients of the non-linear oil film forces and for identifying the magnitude as well as location of unbalance. First of all, the multi-degree-of-freedom equation of motion of the rotor-bearing system, under the action of non-linear oil forces, are derived by Ritz＇ s method according to d＇ Alembert＇ s principle. The question concerning the identification of the non-linear oil film force＇ s parameters and the magnitude of unbalance is then turned into a multi-parametric optimization problem. The parameters and the unbalance are finally identified with the help of genetic algorithm. Results show that by this method the unbalance can effectively be identified with relative high precision, without requiring repeated trial balancing. Figs 5, table 1 and refs 6.