摘要
建立打浆度、定量、水分等信号随机分量的自回归(AR)模型,并采用Levinson-Durb in递推算法对AR模型参数进行在线估计,由AR模型参数得到信号的功率谱估计,并根据功率谱的变化预测断纸过程的发生。将这种方法应用于生产,实践表明预报效果显著。
A new method that predicts paper breaking based on differentiating between stationary property and nonstationary property of some random signals including beating degree,basis weight and moisture is presented. In prediction process, the autoregression (AR) models of these random signals are estimated by Levinson-Durbin recursive algorithm, and power spectral estimation is made by the AR model on line so as to paper breaking can be judged in time from variation of the power spectral. The application results show that the predictive validity is remarkable.
出处
《化工自动化及仪表》
CAS
2007年第2期13-16,共4页
Control and Instruments in Chemical Industry
基金
国家自然科学基金资助项目(6050201)
陕西省教育厅专项科研项目(06JK356)
关键词
断纸
非平稳特性
AR模型
随机信号
功率谱估计
paper breaking
non-stationary property
AR model
random signals
power spectral estimation