摘要
为了提高纸页定量和水分的控制质量 ,本文讨论了利用扫描仪的扫描数据对纸页定量和水分进行在线辨识的横向 (CD)算法和纵向 (MD)算法。针对水分所固有的非线性 ,水分模型 (EIMC)采用了自举算法 ;对于定量 ,利用其模型的线性 ,开发了一个扩展卡尔曼滤波器 (IBPAM) ,来辨识更复杂的纵向动态特性。两种算法都进行了仿真 ,结果表明 :从预测偏差的平方和及残数的空白度来看 ,二阶的ARMA模型会产生最佳的数据适应度。
In order to improve the control quality of paper basis weight and moisture content, the on line identification of basis weight and moisture content in paper machines has been discussed, and algorithms for separating cross machine (CD) and machine direction (MD) variations using scanned data have been proposed. In view of its inherent nonlinearity, the moisture scheme (EIMC) uses a bootstrap algorithm. For basis weight, we can take advantage of the model linearity to develop an extended Kaman filter (IBPAM) to estimate the more complicated MD dynamics. Both algorithms have been simulated. The result shows that a second order ARMA model gives the best fit to the data in terms of sum of squares of prediction errors and in terms of the whiteness of the residual.
出处
《计算机与应用化学》
CAS
CSCD
2000年第3期227-232,共6页
Computers and Applied Chemistry