论文部分内容阅读
提出了一种门限自回归(AR)模型的盲辨识算法,并与常用方法进行比较分析。该算法的特点在保证辨识精度上可大大提高其运行速度,而且阶次越高,该算法的优势越明显。将该方法与隐Markov模型结合,以门限自回归模型各区间的AR子模型系数作为特征向量,以隐Markov模型作为分类器,应用到旋转机械升降速过程的故障诊断中。实验结果表明,这种方法有很好的实用性。
A blind identification algorithm based on threshold autoregressive (AR) model is proposed and compared with the commonly used methods. The characteristics of the algorithm can greatly improve the speed of its identification in ensuring the accuracy of identification, and the higher the order, the more obvious the advantages of the algorithm. This method is combined with hidden Markov model, and the AR sub-model coefficients in each interval of the threshold autoregressive model are taken as eigenvectors. Hidden Markov model is used as a classifier in the fault diagnosis of rotating machinery during the process of raising and lowering speed. Experimental results show that this method has good practicality.