提出了一种基于CHMM的工业回转窑喂煤量变化趋势预测的辅助控制系统方法。采用PCA方法对现场采集的数据进行主成分分析方法进行数据降维,并以主元特征序列作为CHMM模型的观测序列。最后,采用回转窑现场热工数据作为实验数据进行喂煤量变化趋势仿真试验,并以此来实现对专家系统的辅助控制,试验结果表明该方法是有效的。
A CHMM-based trend prediction method of kiln feeding coal is presented in this paper. PCA was employed to extract features of time series by reducing the correlated variables to principal components. The specific data of the rotary kiln thermal data has been used as experiment data. And use it as an aid-control to expert system. Simulation and experimental results show that the proposed method is feasible.