论文部分内容阅读
为了提高动态过程运行状态在线监控效率,提出了基于小波重构与支持向量(support vector machine,SVM)-反向传播神经网络(back propagation neural network,BPNN)相结合的在线智能监控方法.首先,运用离散小波变换对动态过程实测数据流进行重构,并提取其形状特征.其次,利用训练好的小波重构特征的SVM、均值特征的BPNN及重构后形状特征的SVM,对“监控窗口”内实测数据流进行异常模式识别.最后,应用该方法对某精密轴加工过程进行在线智能监控.结果表明:所提模型识别精度高、训练耗时少,其整体性能明显优于小波重构的BPNN模型与基于统计和形状特征的多分类支持向量机(multi-class support vector machine,MSVM)模型,是一种更为有效的动态过程在线智能监控方法.
In order to improve the on-line monitoring efficiency of dynamic process running, an on-line intelligent monitoring method based on wavelet reconstruction and support vector machine (SVM) -backpropagation neural network (BPNN) is proposed.Firstly, The discrete wavelet transform (DWT) is used to reconstruct the measured data stream of the dynamic process and extract its shape features.Secondly, using the trained wavelet to reconstruct the feature SVM, the mean feature BPNN and the reconstructed shape feature SVM, Window "is used to detect the abnormal pattern.Finally, this method is used to monitor the machining process of a precision shaft online.The results show that the proposed model has high recognition accuracy and less training time, and its overall performance is obviously better than wavelet The reconstructed BPNN model and the multi-class support vector machine (MSVM) model based on statistical and shape features are a more effective online intelligent monitoring method for dynamic process.