摘要
文章提出一类基于时域分段线性多项式函数(PLPF)的辨识方法用于大型汽轮机的数学建模和参数辨识。PLPF是一种将传递函数模型经微分方程转化为多重积分,并利用动态响应的离散观察数据构造拟合函数进行分段处理的方法。根据系统模型输入输出时域动态响应数据,采用该方法直接进行有关参数的辨识,建立具有明确参数物理意义的连续时间模型,方法简洁快速准确。该方法成功应用于实际系统的建模和参数辨识,在多个厂站进行了参数辨识的工业试验。
This study introduces a method of modeling and parameter-identification over large turbine based on PLPF in time domain. PLPF is a method of solving multi-integral problem in model function fitting through piecewise linearization after turning transfer function into differential equations. The method builds mathematical models, and identifies parameters according to the input-output sampling data under the disturbance dynamics. The models are continuous time ones with specific physical significance on parameter. The method has the features of simplicity, fast and accuracy. It has been successfully applied into the research of modeling and parameter identification on practical power system. The field tests were carried, out on several plants, and plenty of experience and first-hand data have been obtained. The results verify the validity of the method.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2003年第4期128-133,共6页
Proceedings of the CSEE
