【摘要】 基于ＵＣ轧机板形控制的原理，提出了一种新的采用正交多项式对冷轧薄带板形检测信号进行辨识和分解的数学模型。该模型的特点是在分解过程中避开了三次项的影响，因此，与国外学者所用的勒让德多项式和国内用的一般多项式相比，该模型对板形检测信号进行分解更能反映板形缺陷的实际情况，并便于确定板形控制的执行机构。更多还原

【Abstract】 In this paper,a new mathematical model which uses orthogonal polynomial to identify and decompose the measured signals of thin stirp flatness in cold rolling is given out according to the principle of flatness controlling on UC mill.The characteristic of the model is that it avoids the effect of the third order on the other orders in decomposition,therefor,it can more accurately describe the real situation of the flatness when being used to decompose the measured signals of the flatness compared with the Legendre polynomial used by foreigner scholars and polynomial regression method used by other scholars,and it is convenient for selecting the actuators.更多还原