基于最小均方自适应滤波器的热工过程建模方法  

Modelling Method of Thermal Process Based on Least Mean Square Adaptive Filter

作　　者：刘吉臻[1] 朱红路[1] 常太华[1] 田亮[1] 高伟 童文洪 

机构地区：[1]华北电力大学自动化系,北京市昌平区102206 [2]福建华电可门发电有限公司,福建省福州市350512

出　　处：《中国电机工程学报》2010年第5期107-112,共6页Proceedings of the CSEE

基　　金：国家自然科学基金项目(50776030);国家863高技术基金项目(2007AA04Z163)~~

摘　　要：针对热工对象的时变性特点及其在运行过程中易受到不确定性干扰的影响,提出一种基于最小均方(least mean square,LMS)自适应滤波器的热工过程建模方法。LMS滤波器以未知对象的输入和输出作为激励和期望信号,通过最速下降法得到未知对象的有限脉冲响应(finity impluse response,FIR)模型,其与差分方程或传递函数是等价的。实验仿真和某电厂实际运行数据验证了该算法的有效性。这种建模方法避免了复杂的机理分析,其抽头权值的分布可以表征热工对象的动态特性,为分析热工对象提供了一种手段。For the thermal process is of special features,and it can be disturbed by uncertain factors easily,a novel modelling method based on least mean square （LMS） adaptive filter was presented.Take the input and output of the unknown object as the inspire signal and the destination signal,the LMS filter can give a finity impulse response （FIR） model by steepest descent algorithm.The FIR model is equal to the differential equation and the transform function.The simulation and the test using the data from the power unit show the effectiveness of the method.It avoids the complicated mechanism analysis for the thermal modelling.The distributing of weight exhibits the dynamic feature of the thermal object,and provides a new approach to analyze the thermal object.

关 键 词：热工过程 在线建模 最小均方算法 自适应滤波器 

分 类 号：TK314[动力工程及工程热物理—热能工程] TN24[电子电信—物理电子学]