|
| 摘要: |
| 提出在同步发电机短路试验中应用Prony法实现参数辨识。用Prony法对三相短路电流实测数据进行指数函数的最小二乘拟合,可确定同步发电机的大部分参数。同步发电机参数辨识分4步实现:确定Prony法中的拟合模型阶数、采样总数和采样间隔;对空载三相短路电流进行实测或录波;利用Prony法辨识三相短路电流中各分量对应的特征量;利用所得特征量辨识同步发电机参数。通过对算例发电机进行数值辨识,结果表明该方法的有效性。研究结果显示:取采样时间间隔约为2ms时,辨识效果好;辨识精度与算法中拟合模型阶数、采样时间间隔和采样总数有关,并受信号中噪声的影响,需加以改进。 |
| 关键词: 同步发电机,Prony算法,参数辨识 |
| DOI： |
| 分类号:TM341 |
| 基金项目: |
|
| Application of Prony method to parameter identification of synchronous generators |
|
SU Xiao-lin ZHOU Shuang-xi
|
| Abstract: |
| The Prony method is presented to identify parameters of synchronous generators during their three-phase short-circuit tests.Prony method is used to fit the measured data of three-phase short-circuit currents with an exponential function based on the least square error,which can determine most parameters of the tested synchronous generator.The identification process includes four steps:determine the order of the fitting model,sample number and sample interval of Prony method;measure or record three-phase short-circuit currents without load;identify corresponding characteristic components in measured currents with Prony method;identify parameters of synchronous generators according to identified characteristic components.An example generator is identified and its results demonstrate the effectiveness of the presented method.The study shows that,higher accuracy can be achieved when sample interval is about 2 ms.The identification accuracy is related to the order of the fitting model,sample interval and sample number,and also affected by signal noise,which shows the method should be improved. |
| Key words: synchronous generator,Prony algorithm,parameter identification |
