含有非线性环节的发电机励磁系统参数辨识

提出了一种基于遗传算法的励磁系统参数辨识方法,通过建立待辨识励磁系统的传递函数 结构模型,以励磁系统的实际输入作为模型的输入,以实际励磁系统和模型的输出误差最小作为目 标,利用遗传算法对模型参数进行优化调整,最终得到满足误差要求的励磁系统参数。该方法的优 点在于解决了目前电力系统中常用的辨识方法无法对非线性环节进行有效辨识的问题;且根据输 入输出采样数据直接在时域上进行参数辨识,方法简便,直接得到传递函数框图环节参数,无需转 换。在MATLAB下的数字仿真和现场试验结果均表明,该算法能较精确地辨识出包括非线性环 节在内的励磁系统模型各个环节的参数。

Nonlinear Parameters Identification for Synchronous Generator Excitation Systems

A new method based on genetic algorithm is presented for identifying the excitation system parameters. By building transfer function structure model of the generator excitation system and using the real input data as the input of model, with the minimized difference between output of the model and the real system being the object, the parameters satisfying error request can be finally obtained by adjusting the model parameter with genetic algorithm. The method solves the problem that traditional identification methods such as frequency-domain methods and time-domain methods cannot validly identify the parameters of nonlinear systems. Moreover, according to input and output data, the method identifies parameters in time domain directly and obtains parameters of each transfer function block without conversion operations. The identification results of both a simulation system in MATLAB and a real excitation system show that the new method is able to obtain the accurate parameters of the excitation system consisting nonlinear units.